End of training
Browse files- all_results.json +11 -11
- eval_results.json +6 -6
- runs/Dec18_11-49-22_61f348adfad0/events.out.tfevents.1702901007.61f348adfad0.741.2 +3 -0
- train_results.json +6 -6
- trainer_state.json +173 -176
all_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch": 4.
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
-
"total_flos": 1.
|
| 9 |
-
"train_loss": 0.
|
| 10 |
-
"train_runtime":
|
| 11 |
-
"train_samples_per_second":
|
| 12 |
-
"train_steps_per_second": 0.
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 4.99,
|
| 3 |
+
"eval_accuracy": 0.9796511627906976,
|
| 4 |
+
"eval_loss": 0.07256749272346497,
|
| 5 |
+
"eval_runtime": 7.0579,
|
| 6 |
+
"eval_samples_per_second": 146.218,
|
| 7 |
+
"eval_steps_per_second": 4.676,
|
| 8 |
+
"total_flos": 1.151826529604567e+18,
|
| 9 |
+
"train_loss": 0.13382491601838006,
|
| 10 |
+
"train_runtime": 783.9355,
|
| 11 |
+
"train_samples_per_second": 59.195,
|
| 12 |
+
"train_steps_per_second": 0.459
|
| 13 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch": 4.
|
| 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": 4.99,
|
| 3 |
+
"eval_accuracy": 0.9796511627906976,
|
| 4 |
+
"eval_loss": 0.07256749272346497,
|
| 5 |
+
"eval_runtime": 7.0579,
|
| 6 |
+
"eval_samples_per_second": 146.218,
|
| 7 |
+
"eval_steps_per_second": 4.676
|
| 8 |
}
|
runs/Dec18_11-49-22_61f348adfad0/events.out.tfevents.1702901007.61f348adfad0.741.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:48270b8e93dc15081da5a1919720ac014262971b9532e657996bc8b891714e2b
|
| 3 |
+
size 411
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch": 4.
|
| 3 |
-
"total_flos": 1.
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 4.99,
|
| 3 |
+
"total_flos": 1.151826529604567e+18,
|
| 4 |
+
"train_loss": 0.13382491601838006,
|
| 5 |
+
"train_runtime": 783.9355,
|
| 6 |
+
"train_samples_per_second": 59.195,
|
| 7 |
+
"train_steps_per_second": 0.459
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,8 +1,7 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "swin-tiny-patch4-window7-224-blank_img/checkpoint-
|
| 4 |
-
"epoch": 4.
|
| 5 |
-
"eval_steps": 500,
|
| 6 |
"global_step": 360,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
|
@@ -11,495 +10,493 @@
|
|
| 11 |
{
|
| 12 |
"epoch": 0.07,
|
| 13 |
"learning_rate": 6.944444444444445e-06,
|
| 14 |
-
"loss": 0.
|
| 15 |
"step": 5
|
| 16 |
},
|
| 17 |
{
|
| 18 |
"epoch": 0.14,
|
| 19 |
"learning_rate": 1.388888888888889e-05,
|
| 20 |
-
"loss": 0.
|
| 21 |
"step": 10
|
| 22 |
},
|
| 23 |
{
|
| 24 |
"epoch": 0.21,
|
| 25 |
"learning_rate": 2.0833333333333336e-05,
|
| 26 |
-
"loss": 0.
|
| 27 |
"step": 15
|
| 28 |
},
|
| 29 |
{
|
| 30 |
"epoch": 0.27,
|
| 31 |
"learning_rate": 2.777777777777778e-05,
|
| 32 |
-
"loss": 0.
|
| 33 |
"step": 20
|
| 34 |
},
|
| 35 |
{
|
| 36 |
"epoch": 0.34,
|
| 37 |
"learning_rate": 3.472222222222222e-05,
|
| 38 |
-
"loss": 0.
|
| 39 |
"step": 25
|
| 40 |
},
|
| 41 |
{
|
| 42 |
"epoch": 0.41,
|
| 43 |
"learning_rate": 4.166666666666667e-05,
|
| 44 |
-
"loss": 0.
|
| 45 |
"step": 30
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 0.48,
|
| 49 |
"learning_rate": 4.8611111111111115e-05,
|
| 50 |
-
"loss": 0.
|
| 51 |
"step": 35
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.55,
|
| 55 |
"learning_rate": 4.938271604938271e-05,
|
| 56 |
-
"loss": 0.
|
| 57 |
"step": 40
|
| 58 |
},
|
| 59 |
{
|
| 60 |
"epoch": 0.62,
|
| 61 |
"learning_rate": 4.8611111111111115e-05,
|
| 62 |
-
"loss": 0.
|
| 63 |
"step": 45
|
| 64 |
},
|
| 65 |
{
|
| 66 |
"epoch": 0.69,
|
| 67 |
"learning_rate": 4.783950617283951e-05,
|
| 68 |
-
"loss": 0.
|
| 69 |
"step": 50
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 0.76,
|
| 73 |
"learning_rate": 4.70679012345679e-05,
|
| 74 |
-
"loss": 0.
|
| 75 |
"step": 55
|
| 76 |
},
|
| 77 |
{
|
| 78 |
"epoch": 0.82,
|
| 79 |
"learning_rate": 4.62962962962963e-05,
|
| 80 |
-
"loss": 0.
|
| 81 |
"step": 60
|
| 82 |
},
|
| 83 |
{
|
| 84 |
"epoch": 0.89,
|
| 85 |
"learning_rate": 4.5524691358024696e-05,
|
| 86 |
-
"loss": 0.
|
| 87 |
"step": 65
|
| 88 |
},
|
| 89 |
{
|
| 90 |
"epoch": 0.96,
|
| 91 |
"learning_rate": 4.4753086419753084e-05,
|
| 92 |
-
"loss": 0.
|
| 93 |
"step": 70
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 0.99,
|
| 97 |
-
"eval_accuracy": 0.
|
| 98 |
-
"eval_loss": 0.
|
| 99 |
-
"eval_runtime":
|
| 100 |
-
"eval_samples_per_second":
|
| 101 |
-
"eval_steps_per_second": 4.
|
| 102 |
"step": 72
|
| 103 |
},
|
| 104 |
{
|
| 105 |
-
"epoch": 1.
|
| 106 |
"learning_rate": 4.3981481481481486e-05,
|
| 107 |
-
"loss": 0.
|
| 108 |
"step": 75
|
| 109 |
},
|
| 110 |
{
|
| 111 |
-
"epoch": 1.
|
| 112 |
"learning_rate": 4.3209876543209875e-05,
|
| 113 |
-
"loss": 0.
|
| 114 |
"step": 80
|
| 115 |
},
|
| 116 |
{
|
| 117 |
-
"epoch": 1.
|
| 118 |
"learning_rate": 4.243827160493827e-05,
|
| 119 |
-
"loss": 0.
|
| 120 |
"step": 85
|
| 121 |
},
|
| 122 |
{
|
| 123 |
-
"epoch": 1.
|
| 124 |
"learning_rate": 4.166666666666667e-05,
|
| 125 |
-
"loss": 0.
|
| 126 |
"step": 90
|
| 127 |
},
|
| 128 |
{
|
| 129 |
-
"epoch": 1.
|
| 130 |
"learning_rate": 4.089506172839506e-05,
|
| 131 |
-
"loss": 0.
|
| 132 |
"step": 95
|
| 133 |
},
|
| 134 |
{
|
| 135 |
-
"epoch": 1.
|
| 136 |
"learning_rate": 4.012345679012346e-05,
|
| 137 |
-
"loss": 0.
|
| 138 |
"step": 100
|
| 139 |
},
|
| 140 |
{
|
| 141 |
-
"epoch": 1.
|
| 142 |
"learning_rate": 3.935185185185186e-05,
|
| 143 |
-
"loss": 0.
|
| 144 |
"step": 105
|
| 145 |
},
|
| 146 |
{
|
| 147 |
-
"epoch": 1.
|
| 148 |
"learning_rate": 3.8580246913580246e-05,
|
| 149 |
-
"loss": 0.
|
| 150 |
"step": 110
|
| 151 |
},
|
| 152 |
{
|
| 153 |
-
"epoch": 1.
|
| 154 |
"learning_rate": 3.780864197530865e-05,
|
| 155 |
-
"loss": 0.
|
| 156 |
"step": 115
|
| 157 |
},
|
| 158 |
{
|
| 159 |
-
"epoch": 1.
|
| 160 |
"learning_rate": 3.7037037037037037e-05,
|
| 161 |
-
"loss": 0.
|
| 162 |
"step": 120
|
| 163 |
},
|
| 164 |
{
|
| 165 |
-
"epoch": 1.
|
| 166 |
"learning_rate": 3.626543209876543e-05,
|
| 167 |
-
"loss": 0.
|
| 168 |
"step": 125
|
| 169 |
},
|
| 170 |
{
|
| 171 |
-
"epoch": 1.
|
| 172 |
"learning_rate": 3.5493827160493834e-05,
|
| 173 |
-
"loss": 0.
|
| 174 |
"step": 130
|
| 175 |
},
|
| 176 |
{
|
| 177 |
-
"epoch": 1.
|
| 178 |
"learning_rate": 3.472222222222222e-05,
|
| 179 |
-
"loss": 0.
|
| 180 |
"step": 135
|
| 181 |
},
|
| 182 |
{
|
| 183 |
-
"epoch": 1.
|
| 184 |
"learning_rate": 3.395061728395062e-05,
|
| 185 |
-
"loss": 0.
|
| 186 |
"step": 140
|
| 187 |
},
|
| 188 |
{
|
| 189 |
"epoch": 1.99,
|
| 190 |
-
"
|
| 191 |
-
"
|
| 192 |
-
"
|
|
|
|
|
|
|
|
|
|
| 193 |
},
|
| 194 |
{
|
| 195 |
-
"epoch":
|
| 196 |
-
"
|
| 197 |
-
"
|
| 198 |
-
"eval_runtime": 7.16,
|
| 199 |
-
"eval_samples_per_second": 144.135,
|
| 200 |
-
"eval_steps_per_second": 4.609,
|
| 201 |
"step": 145
|
| 202 |
},
|
| 203 |
{
|
| 204 |
-
"epoch": 2.
|
| 205 |
"learning_rate": 3.240740740740741e-05,
|
| 206 |
-
"loss": 0.
|
| 207 |
"step": 150
|
| 208 |
},
|
| 209 |
{
|
| 210 |
-
"epoch": 2.
|
| 211 |
"learning_rate": 3.16358024691358e-05,
|
| 212 |
-
"loss": 0.
|
| 213 |
"step": 155
|
| 214 |
},
|
| 215 |
{
|
| 216 |
-
"epoch": 2.
|
| 217 |
"learning_rate": 3.08641975308642e-05,
|
| 218 |
-
"loss": 0.
|
| 219 |
"step": 160
|
| 220 |
},
|
| 221 |
{
|
| 222 |
-
"epoch": 2.
|
| 223 |
"learning_rate": 3.0092592592592593e-05,
|
| 224 |
-
"loss": 0.
|
| 225 |
"step": 165
|
| 226 |
},
|
| 227 |
{
|
| 228 |
-
"epoch": 2.
|
| 229 |
"learning_rate": 2.9320987654320992e-05,
|
| 230 |
-
"loss": 0.
|
| 231 |
"step": 170
|
| 232 |
},
|
| 233 |
{
|
| 234 |
-
"epoch": 2.
|
| 235 |
"learning_rate": 2.8549382716049384e-05,
|
| 236 |
-
"loss": 0.
|
| 237 |
"step": 175
|
| 238 |
},
|
| 239 |
{
|
| 240 |
-
"epoch": 2.
|
| 241 |
"learning_rate": 2.777777777777778e-05,
|
| 242 |
-
"loss": 0.
|
| 243 |
"step": 180
|
| 244 |
},
|
| 245 |
{
|
| 246 |
-
"epoch": 2.
|
| 247 |
"learning_rate": 2.700617283950617e-05,
|
| 248 |
-
"loss": 0.
|
| 249 |
"step": 185
|
| 250 |
},
|
| 251 |
{
|
| 252 |
-
"epoch": 2.
|
| 253 |
"learning_rate": 2.623456790123457e-05,
|
| 254 |
-
"loss": 0.
|
| 255 |
"step": 190
|
| 256 |
},
|
| 257 |
{
|
| 258 |
-
"epoch": 2.
|
| 259 |
"learning_rate": 2.5462962962962965e-05,
|
| 260 |
-
"loss": 0.
|
| 261 |
"step": 195
|
| 262 |
},
|
| 263 |
{
|
| 264 |
-
"epoch": 2.
|
| 265 |
"learning_rate": 2.4691358024691357e-05,
|
| 266 |
-
"loss": 0.
|
| 267 |
"step": 200
|
| 268 |
},
|
| 269 |
{
|
| 270 |
-
"epoch": 2.
|
| 271 |
"learning_rate": 2.3919753086419755e-05,
|
| 272 |
-
"loss": 0.
|
| 273 |
"step": 205
|
| 274 |
},
|
| 275 |
{
|
| 276 |
-
"epoch": 2.
|
| 277 |
"learning_rate": 2.314814814814815e-05,
|
| 278 |
-
"loss": 0.
|
| 279 |
"step": 210
|
| 280 |
},
|
| 281 |
{
|
| 282 |
-
"epoch": 2.
|
| 283 |
"learning_rate": 2.2376543209876542e-05,
|
| 284 |
-
"loss": 0.
|
| 285 |
"step": 215
|
| 286 |
},
|
| 287 |
{
|
| 288 |
-
"epoch":
|
| 289 |
-
"eval_accuracy": 0.
|
| 290 |
-
"eval_loss": 0.
|
| 291 |
-
"eval_runtime":
|
| 292 |
-
"eval_samples_per_second":
|
| 293 |
-
"eval_steps_per_second": 4.
|
| 294 |
-
"step":
|
| 295 |
},
|
| 296 |
{
|
| 297 |
-
"epoch": 3.
|
| 298 |
"learning_rate": 2.1604938271604937e-05,
|
| 299 |
-
"loss": 0.
|
| 300 |
"step": 220
|
| 301 |
},
|
| 302 |
{
|
| 303 |
-
"epoch": 3.
|
| 304 |
"learning_rate": 2.0833333333333336e-05,
|
| 305 |
-
"loss": 0.
|
| 306 |
"step": 225
|
| 307 |
},
|
| 308 |
{
|
| 309 |
-
"epoch": 3.
|
| 310 |
"learning_rate": 2.006172839506173e-05,
|
| 311 |
-
"loss": 0.
|
| 312 |
"step": 230
|
| 313 |
},
|
| 314 |
{
|
| 315 |
-
"epoch": 3.
|
| 316 |
"learning_rate": 1.9290123456790123e-05,
|
| 317 |
-
"loss": 0.
|
| 318 |
"step": 235
|
| 319 |
},
|
| 320 |
{
|
| 321 |
-
"epoch": 3.
|
| 322 |
"learning_rate": 1.8518518518518518e-05,
|
| 323 |
-
"loss": 0.
|
| 324 |
"step": 240
|
| 325 |
},
|
| 326 |
{
|
| 327 |
-
"epoch": 3.
|
| 328 |
"learning_rate": 1.7746913580246917e-05,
|
| 329 |
-
"loss": 0.
|
| 330 |
"step": 245
|
| 331 |
},
|
| 332 |
{
|
| 333 |
-
"epoch": 3.
|
| 334 |
"learning_rate": 1.697530864197531e-05,
|
| 335 |
-
"loss": 0.
|
| 336 |
"step": 250
|
| 337 |
},
|
| 338 |
{
|
| 339 |
-
"epoch": 3.
|
| 340 |
"learning_rate": 1.6203703703703704e-05,
|
| 341 |
-
"loss": 0.
|
| 342 |
"step": 255
|
| 343 |
},
|
| 344 |
{
|
| 345 |
-
"epoch": 3.
|
| 346 |
"learning_rate": 1.54320987654321e-05,
|
| 347 |
-
"loss": 0.
|
| 348 |
"step": 260
|
| 349 |
},
|
| 350 |
{
|
| 351 |
-
"epoch": 3.
|
| 352 |
"learning_rate": 1.4660493827160496e-05,
|
| 353 |
-
"loss": 0.
|
| 354 |
"step": 265
|
| 355 |
},
|
| 356 |
{
|
| 357 |
-
"epoch": 3.
|
| 358 |
"learning_rate": 1.388888888888889e-05,
|
| 359 |
-
"loss": 0.
|
| 360 |
"step": 270
|
| 361 |
},
|
| 362 |
{
|
| 363 |
-
"epoch": 3.
|
| 364 |
"learning_rate": 1.3117283950617285e-05,
|
| 365 |
-
"loss": 0.
|
| 366 |
"step": 275
|
| 367 |
},
|
| 368 |
{
|
| 369 |
-
"epoch": 3.
|
| 370 |
"learning_rate": 1.2345679012345678e-05,
|
| 371 |
-
"loss": 0.
|
| 372 |
"step": 280
|
| 373 |
},
|
| 374 |
{
|
| 375 |
-
"epoch": 3.
|
| 376 |
"learning_rate": 1.1574074074074075e-05,
|
| 377 |
-
"loss": 0.
|
| 378 |
"step": 285
|
| 379 |
},
|
| 380 |
{
|
| 381 |
"epoch": 3.99,
|
| 382 |
-
"
|
| 383 |
-
"
|
| 384 |
-
"
|
|
|
|
|
|
|
|
|
|
| 385 |
},
|
| 386 |
{
|
| 387 |
-
"epoch": 4.
|
| 388 |
-
"
|
| 389 |
-
"
|
| 390 |
-
"
|
| 391 |
-
"eval_samples_per_second": 146.199,
|
| 392 |
-
"eval_steps_per_second": 4.675,
|
| 393 |
-
"step": 291
|
| 394 |
},
|
| 395 |
{
|
| 396 |
-
"epoch": 4.
|
| 397 |
"learning_rate": 1.0030864197530866e-05,
|
| 398 |
-
"loss": 0.
|
| 399 |
"step": 295
|
| 400 |
},
|
| 401 |
{
|
| 402 |
-
"epoch": 4.
|
| 403 |
"learning_rate": 9.259259259259259e-06,
|
| 404 |
-
"loss": 0.
|
| 405 |
"step": 300
|
| 406 |
},
|
| 407 |
{
|
| 408 |
-
"epoch": 4.
|
| 409 |
"learning_rate": 8.487654320987654e-06,
|
| 410 |
-
"loss": 0.
|
| 411 |
"step": 305
|
| 412 |
},
|
| 413 |
{
|
| 414 |
-
"epoch": 4.
|
| 415 |
"learning_rate": 7.71604938271605e-06,
|
| 416 |
-
"loss": 0.
|
| 417 |
"step": 310
|
| 418 |
},
|
| 419 |
{
|
| 420 |
-
"epoch": 4.
|
| 421 |
"learning_rate": 6.944444444444445e-06,
|
| 422 |
-
"loss": 0.
|
| 423 |
"step": 315
|
| 424 |
},
|
| 425 |
{
|
| 426 |
-
"epoch": 4.
|
| 427 |
"learning_rate": 6.172839506172839e-06,
|
| 428 |
-
"loss": 0.
|
| 429 |
"step": 320
|
| 430 |
},
|
| 431 |
{
|
| 432 |
-
"epoch": 4.
|
| 433 |
"learning_rate": 5.401234567901234e-06,
|
| 434 |
-
"loss": 0.
|
| 435 |
"step": 325
|
| 436 |
},
|
| 437 |
{
|
| 438 |
-
"epoch": 4.
|
| 439 |
"learning_rate": 4.6296296296296296e-06,
|
| 440 |
-
"loss": 0.
|
| 441 |
"step": 330
|
| 442 |
},
|
| 443 |
{
|
| 444 |
-
"epoch": 4.
|
| 445 |
"learning_rate": 3.858024691358025e-06,
|
| 446 |
-
"loss": 0.
|
| 447 |
"step": 335
|
| 448 |
},
|
| 449 |
{
|
| 450 |
-
"epoch": 4.
|
| 451 |
"learning_rate": 3.0864197530864196e-06,
|
| 452 |
-
"loss": 0.
|
| 453 |
"step": 340
|
| 454 |
},
|
| 455 |
{
|
| 456 |
-
"epoch": 4.
|
| 457 |
"learning_rate": 2.3148148148148148e-06,
|
| 458 |
-
"loss": 0.
|
| 459 |
"step": 345
|
| 460 |
},
|
| 461 |
{
|
| 462 |
-
"epoch": 4.
|
| 463 |
"learning_rate": 1.5432098765432098e-06,
|
| 464 |
-
"loss": 0.
|
| 465 |
"step": 350
|
| 466 |
},
|
| 467 |
{
|
| 468 |
-
"epoch": 4.
|
| 469 |
"learning_rate": 7.716049382716049e-07,
|
| 470 |
-
"loss": 0.
|
| 471 |
"step": 355
|
| 472 |
},
|
| 473 |
{
|
| 474 |
-
"epoch": 4.
|
| 475 |
"learning_rate": 0.0,
|
| 476 |
-
"loss": 0.
|
| 477 |
"step": 360
|
| 478 |
},
|
| 479 |
{
|
| 480 |
-
"epoch": 4.
|
| 481 |
-
"eval_accuracy": 0.
|
| 482 |
-
"eval_loss": 0.
|
| 483 |
-
"eval_runtime": 6.
|
| 484 |
-
"eval_samples_per_second":
|
| 485 |
-
"eval_steps_per_second": 4.
|
| 486 |
"step": 360
|
| 487 |
},
|
| 488 |
{
|
| 489 |
-
"epoch": 4.
|
| 490 |
"step": 360,
|
| 491 |
-
"total_flos": 1.
|
| 492 |
-
"train_loss": 0.
|
| 493 |
-
"train_runtime":
|
| 494 |
-
"train_samples_per_second":
|
| 495 |
-
"train_steps_per_second": 0.
|
| 496 |
}
|
| 497 |
],
|
| 498 |
-
"logging_steps": 5,
|
| 499 |
"max_steps": 360,
|
| 500 |
"num_train_epochs": 5,
|
| 501 |
-
"
|
| 502 |
-
"total_flos": 1.1422818298339983e+18,
|
| 503 |
"trial_name": null,
|
| 504 |
"trial_params": null
|
| 505 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.9796511627906976,
|
| 3 |
+
"best_model_checkpoint": "swin-tiny-patch4-window7-224-blank_img/checkpoint-288",
|
| 4 |
+
"epoch": 4.989690721649485,
|
|
|
|
| 5 |
"global_step": 360,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
|
|
|
| 10 |
{
|
| 11 |
"epoch": 0.07,
|
| 12 |
"learning_rate": 6.944444444444445e-06,
|
| 13 |
+
"loss": 0.7256,
|
| 14 |
"step": 5
|
| 15 |
},
|
| 16 |
{
|
| 17 |
"epoch": 0.14,
|
| 18 |
"learning_rate": 1.388888888888889e-05,
|
| 19 |
+
"loss": 0.6125,
|
| 20 |
"step": 10
|
| 21 |
},
|
| 22 |
{
|
| 23 |
"epoch": 0.21,
|
| 24 |
"learning_rate": 2.0833333333333336e-05,
|
| 25 |
+
"loss": 0.4648,
|
| 26 |
"step": 15
|
| 27 |
},
|
| 28 |
{
|
| 29 |
"epoch": 0.27,
|
| 30 |
"learning_rate": 2.777777777777778e-05,
|
| 31 |
+
"loss": 0.3041,
|
| 32 |
"step": 20
|
| 33 |
},
|
| 34 |
{
|
| 35 |
"epoch": 0.34,
|
| 36 |
"learning_rate": 3.472222222222222e-05,
|
| 37 |
+
"loss": 0.2065,
|
| 38 |
"step": 25
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 0.41,
|
| 42 |
"learning_rate": 4.166666666666667e-05,
|
| 43 |
+
"loss": 0.1469,
|
| 44 |
"step": 30
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.48,
|
| 48 |
"learning_rate": 4.8611111111111115e-05,
|
| 49 |
+
"loss": 0.1734,
|
| 50 |
"step": 35
|
| 51 |
},
|
| 52 |
{
|
| 53 |
"epoch": 0.55,
|
| 54 |
"learning_rate": 4.938271604938271e-05,
|
| 55 |
+
"loss": 0.1736,
|
| 56 |
"step": 40
|
| 57 |
},
|
| 58 |
{
|
| 59 |
"epoch": 0.62,
|
| 60 |
"learning_rate": 4.8611111111111115e-05,
|
| 61 |
+
"loss": 0.1559,
|
| 62 |
"step": 45
|
| 63 |
},
|
| 64 |
{
|
| 65 |
"epoch": 0.69,
|
| 66 |
"learning_rate": 4.783950617283951e-05,
|
| 67 |
+
"loss": 0.1355,
|
| 68 |
"step": 50
|
| 69 |
},
|
| 70 |
{
|
| 71 |
"epoch": 0.76,
|
| 72 |
"learning_rate": 4.70679012345679e-05,
|
| 73 |
+
"loss": 0.0896,
|
| 74 |
"step": 55
|
| 75 |
},
|
| 76 |
{
|
| 77 |
"epoch": 0.82,
|
| 78 |
"learning_rate": 4.62962962962963e-05,
|
| 79 |
+
"loss": 0.0994,
|
| 80 |
"step": 60
|
| 81 |
},
|
| 82 |
{
|
| 83 |
"epoch": 0.89,
|
| 84 |
"learning_rate": 4.5524691358024696e-05,
|
| 85 |
+
"loss": 0.127,
|
| 86 |
"step": 65
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 0.96,
|
| 90 |
"learning_rate": 4.4753086419753084e-05,
|
| 91 |
+
"loss": 0.1329,
|
| 92 |
"step": 70
|
| 93 |
},
|
| 94 |
{
|
| 95 |
"epoch": 0.99,
|
| 96 |
+
"eval_accuracy": 0.9767441860465116,
|
| 97 |
+
"eval_loss": 0.08817464113235474,
|
| 98 |
+
"eval_runtime": 7.3528,
|
| 99 |
+
"eval_samples_per_second": 140.354,
|
| 100 |
+
"eval_steps_per_second": 4.488,
|
| 101 |
"step": 72
|
| 102 |
},
|
| 103 |
{
|
| 104 |
+
"epoch": 1.04,
|
| 105 |
"learning_rate": 4.3981481481481486e-05,
|
| 106 |
+
"loss": 0.1512,
|
| 107 |
"step": 75
|
| 108 |
},
|
| 109 |
{
|
| 110 |
+
"epoch": 1.11,
|
| 111 |
"learning_rate": 4.3209876543209875e-05,
|
| 112 |
+
"loss": 0.1248,
|
| 113 |
"step": 80
|
| 114 |
},
|
| 115 |
{
|
| 116 |
+
"epoch": 1.18,
|
| 117 |
"learning_rate": 4.243827160493827e-05,
|
| 118 |
+
"loss": 0.1396,
|
| 119 |
"step": 85
|
| 120 |
},
|
| 121 |
{
|
| 122 |
+
"epoch": 1.25,
|
| 123 |
"learning_rate": 4.166666666666667e-05,
|
| 124 |
+
"loss": 0.1231,
|
| 125 |
"step": 90
|
| 126 |
},
|
| 127 |
{
|
| 128 |
+
"epoch": 1.32,
|
| 129 |
"learning_rate": 4.089506172839506e-05,
|
| 130 |
+
"loss": 0.1228,
|
| 131 |
"step": 95
|
| 132 |
},
|
| 133 |
{
|
| 134 |
+
"epoch": 1.38,
|
| 135 |
"learning_rate": 4.012345679012346e-05,
|
| 136 |
+
"loss": 0.1277,
|
| 137 |
"step": 100
|
| 138 |
},
|
| 139 |
{
|
| 140 |
+
"epoch": 1.45,
|
| 141 |
"learning_rate": 3.935185185185186e-05,
|
| 142 |
+
"loss": 0.1116,
|
| 143 |
"step": 105
|
| 144 |
},
|
| 145 |
{
|
| 146 |
+
"epoch": 1.52,
|
| 147 |
"learning_rate": 3.8580246913580246e-05,
|
| 148 |
+
"loss": 0.1236,
|
| 149 |
"step": 110
|
| 150 |
},
|
| 151 |
{
|
| 152 |
+
"epoch": 1.59,
|
| 153 |
"learning_rate": 3.780864197530865e-05,
|
| 154 |
+
"loss": 0.1254,
|
| 155 |
"step": 115
|
| 156 |
},
|
| 157 |
{
|
| 158 |
+
"epoch": 1.66,
|
| 159 |
"learning_rate": 3.7037037037037037e-05,
|
| 160 |
+
"loss": 0.1004,
|
| 161 |
"step": 120
|
| 162 |
},
|
| 163 |
{
|
| 164 |
+
"epoch": 1.73,
|
| 165 |
"learning_rate": 3.626543209876543e-05,
|
| 166 |
+
"loss": 0.1199,
|
| 167 |
"step": 125
|
| 168 |
},
|
| 169 |
{
|
| 170 |
+
"epoch": 1.8,
|
| 171 |
"learning_rate": 3.5493827160493834e-05,
|
| 172 |
+
"loss": 0.1098,
|
| 173 |
"step": 130
|
| 174 |
},
|
| 175 |
{
|
| 176 |
+
"epoch": 1.87,
|
| 177 |
"learning_rate": 3.472222222222222e-05,
|
| 178 |
+
"loss": 0.0787,
|
| 179 |
"step": 135
|
| 180 |
},
|
| 181 |
{
|
| 182 |
+
"epoch": 1.93,
|
| 183 |
"learning_rate": 3.395061728395062e-05,
|
| 184 |
+
"loss": 0.1247,
|
| 185 |
"step": 140
|
| 186 |
},
|
| 187 |
{
|
| 188 |
"epoch": 1.99,
|
| 189 |
+
"eval_accuracy": 0.9767441860465116,
|
| 190 |
+
"eval_loss": 0.08047417551279068,
|
| 191 |
+
"eval_runtime": 7.6368,
|
| 192 |
+
"eval_samples_per_second": 135.135,
|
| 193 |
+
"eval_steps_per_second": 4.321,
|
| 194 |
+
"step": 144
|
| 195 |
},
|
| 196 |
{
|
| 197 |
+
"epoch": 2.01,
|
| 198 |
+
"learning_rate": 3.317901234567901e-05,
|
| 199 |
+
"loss": 0.1116,
|
|
|
|
|
|
|
|
|
|
| 200 |
"step": 145
|
| 201 |
},
|
| 202 |
{
|
| 203 |
+
"epoch": 2.08,
|
| 204 |
"learning_rate": 3.240740740740741e-05,
|
| 205 |
+
"loss": 0.122,
|
| 206 |
"step": 150
|
| 207 |
},
|
| 208 |
{
|
| 209 |
+
"epoch": 2.15,
|
| 210 |
"learning_rate": 3.16358024691358e-05,
|
| 211 |
+
"loss": 0.1109,
|
| 212 |
"step": 155
|
| 213 |
},
|
| 214 |
{
|
| 215 |
+
"epoch": 2.22,
|
| 216 |
"learning_rate": 3.08641975308642e-05,
|
| 217 |
+
"loss": 0.109,
|
| 218 |
"step": 160
|
| 219 |
},
|
| 220 |
{
|
| 221 |
+
"epoch": 2.29,
|
| 222 |
"learning_rate": 3.0092592592592593e-05,
|
| 223 |
+
"loss": 0.104,
|
| 224 |
"step": 165
|
| 225 |
},
|
| 226 |
{
|
| 227 |
+
"epoch": 2.36,
|
| 228 |
"learning_rate": 2.9320987654320992e-05,
|
| 229 |
+
"loss": 0.0849,
|
| 230 |
"step": 170
|
| 231 |
},
|
| 232 |
{
|
| 233 |
+
"epoch": 2.43,
|
| 234 |
"learning_rate": 2.8549382716049384e-05,
|
| 235 |
+
"loss": 0.1088,
|
| 236 |
"step": 175
|
| 237 |
},
|
| 238 |
{
|
| 239 |
+
"epoch": 2.49,
|
| 240 |
"learning_rate": 2.777777777777778e-05,
|
| 241 |
+
"loss": 0.0988,
|
| 242 |
"step": 180
|
| 243 |
},
|
| 244 |
{
|
| 245 |
+
"epoch": 2.56,
|
| 246 |
"learning_rate": 2.700617283950617e-05,
|
| 247 |
+
"loss": 0.1072,
|
| 248 |
"step": 185
|
| 249 |
},
|
| 250 |
{
|
| 251 |
+
"epoch": 2.63,
|
| 252 |
"learning_rate": 2.623456790123457e-05,
|
| 253 |
+
"loss": 0.1098,
|
| 254 |
"step": 190
|
| 255 |
},
|
| 256 |
{
|
| 257 |
+
"epoch": 2.7,
|
| 258 |
"learning_rate": 2.5462962962962965e-05,
|
| 259 |
+
"loss": 0.1288,
|
| 260 |
"step": 195
|
| 261 |
},
|
| 262 |
{
|
| 263 |
+
"epoch": 2.77,
|
| 264 |
"learning_rate": 2.4691358024691357e-05,
|
| 265 |
+
"loss": 0.1401,
|
| 266 |
"step": 200
|
| 267 |
},
|
| 268 |
{
|
| 269 |
+
"epoch": 2.84,
|
| 270 |
"learning_rate": 2.3919753086419755e-05,
|
| 271 |
+
"loss": 0.1237,
|
| 272 |
"step": 205
|
| 273 |
},
|
| 274 |
{
|
| 275 |
+
"epoch": 2.91,
|
| 276 |
"learning_rate": 2.314814814814815e-05,
|
| 277 |
+
"loss": 0.1063,
|
| 278 |
"step": 210
|
| 279 |
},
|
| 280 |
{
|
| 281 |
+
"epoch": 2.98,
|
| 282 |
"learning_rate": 2.2376543209876542e-05,
|
| 283 |
+
"loss": 0.0742,
|
| 284 |
"step": 215
|
| 285 |
},
|
| 286 |
{
|
| 287 |
+
"epoch": 2.99,
|
| 288 |
+
"eval_accuracy": 0.9767441860465116,
|
| 289 |
+
"eval_loss": 0.07210300117731094,
|
| 290 |
+
"eval_runtime": 7.3216,
|
| 291 |
+
"eval_samples_per_second": 140.953,
|
| 292 |
+
"eval_steps_per_second": 4.507,
|
| 293 |
+
"step": 216
|
| 294 |
},
|
| 295 |
{
|
| 296 |
+
"epoch": 3.05,
|
| 297 |
"learning_rate": 2.1604938271604937e-05,
|
| 298 |
+
"loss": 0.1008,
|
| 299 |
"step": 220
|
| 300 |
},
|
| 301 |
{
|
| 302 |
+
"epoch": 3.12,
|
| 303 |
"learning_rate": 2.0833333333333336e-05,
|
| 304 |
+
"loss": 0.1196,
|
| 305 |
"step": 225
|
| 306 |
},
|
| 307 |
{
|
| 308 |
+
"epoch": 3.19,
|
| 309 |
"learning_rate": 2.006172839506173e-05,
|
| 310 |
+
"loss": 0.0712,
|
| 311 |
"step": 230
|
| 312 |
},
|
| 313 |
{
|
| 314 |
+
"epoch": 3.26,
|
| 315 |
"learning_rate": 1.9290123456790123e-05,
|
| 316 |
+
"loss": 0.0726,
|
| 317 |
"step": 235
|
| 318 |
},
|
| 319 |
{
|
| 320 |
+
"epoch": 3.33,
|
| 321 |
"learning_rate": 1.8518518518518518e-05,
|
| 322 |
+
"loss": 0.0964,
|
| 323 |
"step": 240
|
| 324 |
},
|
| 325 |
{
|
| 326 |
+
"epoch": 3.4,
|
| 327 |
"learning_rate": 1.7746913580246917e-05,
|
| 328 |
+
"loss": 0.0932,
|
| 329 |
"step": 245
|
| 330 |
},
|
| 331 |
{
|
| 332 |
+
"epoch": 3.47,
|
| 333 |
"learning_rate": 1.697530864197531e-05,
|
| 334 |
+
"loss": 0.1022,
|
| 335 |
"step": 250
|
| 336 |
},
|
| 337 |
{
|
| 338 |
+
"epoch": 3.54,
|
| 339 |
"learning_rate": 1.6203703703703704e-05,
|
| 340 |
+
"loss": 0.1033,
|
| 341 |
"step": 255
|
| 342 |
},
|
| 343 |
{
|
| 344 |
+
"epoch": 3.6,
|
| 345 |
"learning_rate": 1.54320987654321e-05,
|
| 346 |
+
"loss": 0.0913,
|
| 347 |
"step": 260
|
| 348 |
},
|
| 349 |
{
|
| 350 |
+
"epoch": 3.67,
|
| 351 |
"learning_rate": 1.4660493827160496e-05,
|
| 352 |
+
"loss": 0.0873,
|
| 353 |
"step": 265
|
| 354 |
},
|
| 355 |
{
|
| 356 |
+
"epoch": 3.74,
|
| 357 |
"learning_rate": 1.388888888888889e-05,
|
| 358 |
+
"loss": 0.1095,
|
| 359 |
"step": 270
|
| 360 |
},
|
| 361 |
{
|
| 362 |
+
"epoch": 3.81,
|
| 363 |
"learning_rate": 1.3117283950617285e-05,
|
| 364 |
+
"loss": 0.1047,
|
| 365 |
"step": 275
|
| 366 |
},
|
| 367 |
{
|
| 368 |
+
"epoch": 3.88,
|
| 369 |
"learning_rate": 1.2345679012345678e-05,
|
| 370 |
+
"loss": 0.0707,
|
| 371 |
"step": 280
|
| 372 |
},
|
| 373 |
{
|
| 374 |
+
"epoch": 3.95,
|
| 375 |
"learning_rate": 1.1574074074074075e-05,
|
| 376 |
+
"loss": 0.0745,
|
| 377 |
"step": 285
|
| 378 |
},
|
| 379 |
{
|
| 380 |
"epoch": 3.99,
|
| 381 |
+
"eval_accuracy": 0.9796511627906976,
|
| 382 |
+
"eval_loss": 0.07256749272346497,
|
| 383 |
+
"eval_runtime": 6.7881,
|
| 384 |
+
"eval_samples_per_second": 152.03,
|
| 385 |
+
"eval_steps_per_second": 4.861,
|
| 386 |
+
"step": 288
|
| 387 |
},
|
| 388 |
{
|
| 389 |
+
"epoch": 4.03,
|
| 390 |
+
"learning_rate": 1.0802469135802469e-05,
|
| 391 |
+
"loss": 0.129,
|
| 392 |
+
"step": 290
|
|
|
|
|
|
|
|
|
|
| 393 |
},
|
| 394 |
{
|
| 395 |
+
"epoch": 4.1,
|
| 396 |
"learning_rate": 1.0030864197530866e-05,
|
| 397 |
+
"loss": 0.1298,
|
| 398 |
"step": 295
|
| 399 |
},
|
| 400 |
{
|
| 401 |
+
"epoch": 4.16,
|
| 402 |
"learning_rate": 9.259259259259259e-06,
|
| 403 |
+
"loss": 0.089,
|
| 404 |
"step": 300
|
| 405 |
},
|
| 406 |
{
|
| 407 |
+
"epoch": 4.23,
|
| 408 |
"learning_rate": 8.487654320987654e-06,
|
| 409 |
+
"loss": 0.0907,
|
| 410 |
"step": 305
|
| 411 |
},
|
| 412 |
{
|
| 413 |
+
"epoch": 4.3,
|
| 414 |
"learning_rate": 7.71604938271605e-06,
|
| 415 |
+
"loss": 0.0983,
|
| 416 |
"step": 310
|
| 417 |
},
|
| 418 |
{
|
| 419 |
+
"epoch": 4.37,
|
| 420 |
"learning_rate": 6.944444444444445e-06,
|
| 421 |
+
"loss": 0.0769,
|
| 422 |
"step": 315
|
| 423 |
},
|
| 424 |
{
|
| 425 |
+
"epoch": 4.44,
|
| 426 |
"learning_rate": 6.172839506172839e-06,
|
| 427 |
+
"loss": 0.1115,
|
| 428 |
"step": 320
|
| 429 |
},
|
| 430 |
{
|
| 431 |
+
"epoch": 4.51,
|
| 432 |
"learning_rate": 5.401234567901234e-06,
|
| 433 |
+
"loss": 0.058,
|
| 434 |
"step": 325
|
| 435 |
},
|
| 436 |
{
|
| 437 |
+
"epoch": 4.58,
|
| 438 |
"learning_rate": 4.6296296296296296e-06,
|
| 439 |
+
"loss": 0.0972,
|
| 440 |
"step": 330
|
| 441 |
},
|
| 442 |
{
|
| 443 |
+
"epoch": 4.65,
|
| 444 |
"learning_rate": 3.858024691358025e-06,
|
| 445 |
+
"loss": 0.1036,
|
| 446 |
"step": 335
|
| 447 |
},
|
| 448 |
{
|
| 449 |
+
"epoch": 4.71,
|
| 450 |
"learning_rate": 3.0864197530864196e-06,
|
| 451 |
+
"loss": 0.0675,
|
| 452 |
"step": 340
|
| 453 |
},
|
| 454 |
{
|
| 455 |
+
"epoch": 4.78,
|
| 456 |
"learning_rate": 2.3148148148148148e-06,
|
| 457 |
+
"loss": 0.0909,
|
| 458 |
"step": 345
|
| 459 |
},
|
| 460 |
{
|
| 461 |
+
"epoch": 4.85,
|
| 462 |
"learning_rate": 1.5432098765432098e-06,
|
| 463 |
+
"loss": 0.1148,
|
| 464 |
"step": 350
|
| 465 |
},
|
| 466 |
{
|
| 467 |
+
"epoch": 4.92,
|
| 468 |
"learning_rate": 7.716049382716049e-07,
|
| 469 |
+
"loss": 0.0802,
|
| 470 |
"step": 355
|
| 471 |
},
|
| 472 |
{
|
| 473 |
+
"epoch": 4.99,
|
| 474 |
"learning_rate": 0.0,
|
| 475 |
+
"loss": 0.1289,
|
| 476 |
"step": 360
|
| 477 |
},
|
| 478 |
{
|
| 479 |
+
"epoch": 4.99,
|
| 480 |
+
"eval_accuracy": 0.9728682170542635,
|
| 481 |
+
"eval_loss": 0.08483530580997467,
|
| 482 |
+
"eval_runtime": 6.8918,
|
| 483 |
+
"eval_samples_per_second": 149.743,
|
| 484 |
+
"eval_steps_per_second": 4.788,
|
| 485 |
"step": 360
|
| 486 |
},
|
| 487 |
{
|
| 488 |
+
"epoch": 4.99,
|
| 489 |
"step": 360,
|
| 490 |
+
"total_flos": 1.151826529604567e+18,
|
| 491 |
+
"train_loss": 0.13382491601838006,
|
| 492 |
+
"train_runtime": 783.9355,
|
| 493 |
+
"train_samples_per_second": 59.195,
|
| 494 |
+
"train_steps_per_second": 0.459
|
| 495 |
}
|
| 496 |
],
|
|
|
|
| 497 |
"max_steps": 360,
|
| 498 |
"num_train_epochs": 5,
|
| 499 |
+
"total_flos": 1.151826529604567e+18,
|
|
|
|
| 500 |
"trial_name": null,
|
| 501 |
"trial_params": null
|
| 502 |
}
|