End of training
Browse files- README.md +2 -2
- all_results.json +11 -11
- eval_results.json +6 -6
- runs/Jan13_19-54-30_a732f801c5be/events.out.tfevents.1705177969.a732f801c5be.827.6 +3 -0
- train_results.json +6 -6
- trainer_state.json +158 -341
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/resnet-101](https://huggingface.co/microsoft/resnet-101) 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/resnet-101](https://huggingface.co/microsoft/resnet-101) on an unknown dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 0.6763
|
| 21 |
+
- Accuracy: 0.7701
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
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": 2.
|
| 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": 9.8,
|
| 3 |
+
"eval_accuracy": 0.7700980392156863,
|
| 4 |
+
"eval_loss": 0.6762834191322327,
|
| 5 |
+
"eval_runtime": 27.2812,
|
| 6 |
+
"eval_samples_per_second": 74.777,
|
| 7 |
+
"eval_steps_per_second": 2.346,
|
| 8 |
+
"total_flos": 3.072509726495183e+18,
|
| 9 |
+
"train_loss": 1.3191132202148437,
|
| 10 |
+
"train_runtime": 2261.7516,
|
| 11 |
+
"train_samples_per_second": 36.074,
|
| 12 |
+
"train_steps_per_second": 0.111
|
| 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": 2.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 9.8,
|
| 3 |
+
"eval_accuracy": 0.7700980392156863,
|
| 4 |
+
"eval_loss": 0.6762834191322327,
|
| 5 |
+
"eval_runtime": 27.2812,
|
| 6 |
+
"eval_samples_per_second": 74.777,
|
| 7 |
+
"eval_steps_per_second": 2.346
|
| 8 |
}
|
runs/Jan13_19-54-30_a732f801c5be/events.out.tfevents.1705177969.a732f801c5be.827.6
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ebd8aaffa626a56c6d41efef37fe5ae608ebdc9d39b31f5edd35f950ed4c1527
|
| 3 |
+
size 411
|
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": 9.8,
|
| 3 |
+
"total_flos": 3.072509726495183e+18,
|
| 4 |
+
"train_loss": 1.3191132202148437,
|
| 5 |
+
"train_runtime": 2261.7516,
|
| 6 |
+
"train_samples_per_second": 36.074,
|
| 7 |
+
"train_steps_per_second": 0.111
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,451 +1,268 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "resnet-101-finetuned-CivilEng11k/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch": 0.
|
| 13 |
-
"learning_rate":
|
| 14 |
-
"loss": 2.
|
| 15 |
-
"step": 5
|
| 16 |
-
},
|
| 17 |
-
{
|
| 18 |
-
"epoch": 0.16,
|
| 19 |
-
"learning_rate": 3.125e-05,
|
| 20 |
-
"loss": 2.6155,
|
| 21 |
"step": 10
|
| 22 |
},
|
| 23 |
{
|
| 24 |
-
"epoch": 0.
|
| 25 |
-
"learning_rate":
|
| 26 |
-
"loss": 2.
|
| 27 |
-
"step": 15
|
| 28 |
-
},
|
| 29 |
-
{
|
| 30 |
-
"epoch": 0.31,
|
| 31 |
-
"learning_rate": 6.25e-05,
|
| 32 |
-
"loss": 2.577,
|
| 33 |
"step": 20
|
| 34 |
},
|
| 35 |
{
|
| 36 |
-
"epoch": 0.
|
| 37 |
-
"
|
| 38 |
-
"
|
|
|
|
|
|
|
|
|
|
| 39 |
"step": 25
|
| 40 |
},
|
| 41 |
{
|
| 42 |
-
"epoch":
|
| 43 |
-
"learning_rate": 9.
|
| 44 |
-
"loss": 2.
|
| 45 |
"step": 30
|
| 46 |
},
|
| 47 |
{
|
| 48 |
-
"epoch":
|
| 49 |
-
"learning_rate": 9.
|
| 50 |
-
"loss": 2.
|
| 51 |
-
"step": 35
|
| 52 |
-
},
|
| 53 |
-
{
|
| 54 |
-
"epoch": 0.63,
|
| 55 |
-
"learning_rate": 9.71731448763251e-05,
|
| 56 |
-
"loss": 2.4167,
|
| 57 |
"step": 40
|
| 58 |
},
|
| 59 |
{
|
| 60 |
-
"epoch":
|
| 61 |
-
"learning_rate":
|
| 62 |
-
"loss": 2.
|
| 63 |
-
"step": 45
|
| 64 |
-
},
|
| 65 |
-
{
|
| 66 |
-
"epoch": 0.78,
|
| 67 |
-
"learning_rate": 9.363957597173145e-05,
|
| 68 |
-
"loss": 2.2503,
|
| 69 |
"step": 50
|
| 70 |
},
|
| 71 |
{
|
| 72 |
-
"epoch": 0
|
| 73 |
-
"
|
| 74 |
-
"
|
| 75 |
-
"
|
|
|
|
|
|
|
|
|
|
| 76 |
},
|
| 77 |
{
|
| 78 |
-
"epoch":
|
| 79 |
-
"learning_rate":
|
| 80 |
-
"loss": 2.
|
| 81 |
"step": 60
|
| 82 |
},
|
| 83 |
{
|
| 84 |
-
"epoch":
|
| 85 |
-
"
|
| 86 |
-
"
|
| 87 |
-
"eval_runtime": 32.3617,
|
| 88 |
-
"eval_samples_per_second": 63.037,
|
| 89 |
-
"eval_steps_per_second": 1.978,
|
| 90 |
-
"step": 63
|
| 91 |
-
},
|
| 92 |
-
{
|
| 93 |
-
"epoch": 1.02,
|
| 94 |
-
"learning_rate": 8.833922261484099e-05,
|
| 95 |
-
"loss": 1.9785,
|
| 96 |
-
"step": 65
|
| 97 |
-
},
|
| 98 |
-
{
|
| 99 |
-
"epoch": 1.1,
|
| 100 |
-
"learning_rate": 8.657243816254418e-05,
|
| 101 |
-
"loss": 1.8484,
|
| 102 |
"step": 70
|
| 103 |
},
|
| 104 |
{
|
| 105 |
-
"epoch":
|
| 106 |
-
"
|
| 107 |
-
"
|
| 108 |
-
"
|
|
|
|
|
|
|
|
|
|
| 109 |
},
|
| 110 |
{
|
| 111 |
-
"epoch":
|
| 112 |
-
"learning_rate":
|
| 113 |
-
"loss": 1.
|
| 114 |
"step": 80
|
| 115 |
},
|
| 116 |
{
|
| 117 |
-
"epoch":
|
| 118 |
-
"learning_rate":
|
| 119 |
-
"loss": 1.
|
| 120 |
-
"step": 85
|
| 121 |
-
},
|
| 122 |
-
{
|
| 123 |
-
"epoch": 1.41,
|
| 124 |
-
"learning_rate": 7.95053003533569e-05,
|
| 125 |
-
"loss": 1.4728,
|
| 126 |
"step": 90
|
| 127 |
},
|
| 128 |
{
|
| 129 |
-
"epoch":
|
| 130 |
-
"learning_rate":
|
| 131 |
-
"loss": 1.
|
| 132 |
-
"step": 95
|
| 133 |
-
},
|
| 134 |
-
{
|
| 135 |
-
"epoch": 1.57,
|
| 136 |
-
"learning_rate": 7.597173144876325e-05,
|
| 137 |
-
"loss": 1.4121,
|
| 138 |
"step": 100
|
| 139 |
},
|
| 140 |
{
|
| 141 |
-
"epoch":
|
| 142 |
-
"
|
| 143 |
-
"
|
| 144 |
-
"
|
|
|
|
|
|
|
|
|
|
| 145 |
},
|
| 146 |
{
|
| 147 |
-
"epoch":
|
| 148 |
-
"learning_rate":
|
| 149 |
-
"loss": 1.
|
| 150 |
"step": 110
|
| 151 |
},
|
| 152 |
{
|
| 153 |
-
"epoch":
|
| 154 |
-
"learning_rate":
|
| 155 |
-
"loss": 1.
|
| 156 |
-
"step": 115
|
| 157 |
-
},
|
| 158 |
-
{
|
| 159 |
-
"epoch": 1.88,
|
| 160 |
-
"learning_rate": 6.890459363957598e-05,
|
| 161 |
-
"loss": 1.3202,
|
| 162 |
"step": 120
|
| 163 |
},
|
| 164 |
{
|
| 165 |
-
"epoch":
|
| 166 |
-
"
|
| 167 |
-
"
|
| 168 |
-
"
|
| 169 |
-
|
| 170 |
-
|
| 171 |
-
"epoch": 1.99,
|
| 172 |
-
"eval_accuracy": 0.5779411764705882,
|
| 173 |
-
"eval_loss": 1.253934621810913,
|
| 174 |
-
"eval_runtime": 35.8869,
|
| 175 |
-
"eval_samples_per_second": 56.845,
|
| 176 |
-
"eval_steps_per_second": 1.783,
|
| 177 |
"step": 127
|
| 178 |
},
|
| 179 |
{
|
| 180 |
-
"epoch":
|
| 181 |
-
"learning_rate":
|
| 182 |
-
"loss": 1.
|
| 183 |
"step": 130
|
| 184 |
},
|
| 185 |
{
|
| 186 |
-
"epoch":
|
| 187 |
-
"learning_rate":
|
| 188 |
-
"loss":
|
| 189 |
-
"step": 135
|
| 190 |
-
},
|
| 191 |
-
{
|
| 192 |
-
"epoch": 2.2,
|
| 193 |
-
"learning_rate": 6.183745583038869e-05,
|
| 194 |
-
"loss": 1.2239,
|
| 195 |
"step": 140
|
| 196 |
},
|
| 197 |
{
|
| 198 |
-
"epoch":
|
| 199 |
-
"learning_rate":
|
| 200 |
-
"loss":
|
| 201 |
-
"step": 145
|
| 202 |
-
},
|
| 203 |
-
{
|
| 204 |
-
"epoch": 2.35,
|
| 205 |
-
"learning_rate": 5.830388692579506e-05,
|
| 206 |
-
"loss": 1.1971,
|
| 207 |
"step": 150
|
| 208 |
},
|
| 209 |
{
|
| 210 |
-
"epoch":
|
| 211 |
-
"
|
| 212 |
-
"
|
| 213 |
-
"
|
|
|
|
|
|
|
|
|
|
| 214 |
},
|
| 215 |
{
|
| 216 |
-
"epoch":
|
| 217 |
-
"learning_rate":
|
| 218 |
-
"loss":
|
| 219 |
"step": 160
|
| 220 |
},
|
| 221 |
{
|
| 222 |
-
"epoch":
|
| 223 |
-
"learning_rate":
|
| 224 |
-
"loss":
|
| 225 |
-
"step": 165
|
| 226 |
-
},
|
| 227 |
-
{
|
| 228 |
-
"epoch": 2.67,
|
| 229 |
-
"learning_rate": 5.123674911660777e-05,
|
| 230 |
-
"loss": 1.1064,
|
| 231 |
"step": 170
|
| 232 |
},
|
| 233 |
{
|
| 234 |
-
"epoch":
|
| 235 |
-
"
|
| 236 |
-
"
|
| 237 |
-
"
|
|
|
|
|
|
|
|
|
|
| 238 |
},
|
| 239 |
{
|
| 240 |
-
"epoch":
|
| 241 |
-
"learning_rate":
|
| 242 |
-
"loss":
|
| 243 |
"step": 180
|
| 244 |
},
|
| 245 |
{
|
| 246 |
-
"epoch":
|
| 247 |
-
"learning_rate":
|
| 248 |
-
"loss":
|
| 249 |
-
"step": 185
|
| 250 |
-
},
|
| 251 |
-
{
|
| 252 |
-
"epoch": 2.98,
|
| 253 |
-
"learning_rate": 4.416961130742049e-05,
|
| 254 |
-
"loss": 1.0809,
|
| 255 |
"step": 190
|
| 256 |
},
|
| 257 |
{
|
| 258 |
-
"epoch":
|
| 259 |
-
"
|
| 260 |
-
"
|
| 261 |
-
"eval_runtime": 38.0303,
|
| 262 |
-
"eval_samples_per_second": 53.641,
|
| 263 |
-
"eval_steps_per_second": 1.683,
|
| 264 |
-
"step": 191
|
| 265 |
-
},
|
| 266 |
-
{
|
| 267 |
-
"epoch": 3.06,
|
| 268 |
-
"learning_rate": 4.240282685512368e-05,
|
| 269 |
-
"loss": 0.9985,
|
| 270 |
-
"step": 195
|
| 271 |
-
},
|
| 272 |
-
{
|
| 273 |
-
"epoch": 3.14,
|
| 274 |
-
"learning_rate": 4.063604240282686e-05,
|
| 275 |
-
"loss": 1.0406,
|
| 276 |
"step": 200
|
| 277 |
},
|
| 278 |
{
|
| 279 |
-
"epoch":
|
| 280 |
-
"
|
| 281 |
-
"
|
| 282 |
-
"
|
|
|
|
|
|
|
|
|
|
| 283 |
},
|
| 284 |
{
|
| 285 |
-
"epoch":
|
| 286 |
-
"learning_rate":
|
| 287 |
-
"loss": 0.
|
| 288 |
"step": 210
|
| 289 |
},
|
| 290 |
{
|
| 291 |
-
"epoch":
|
| 292 |
-
"learning_rate":
|
| 293 |
-
"loss":
|
| 294 |
-
"step": 215
|
| 295 |
-
},
|
| 296 |
-
{
|
| 297 |
-
"epoch": 3.45,
|
| 298 |
-
"learning_rate": 3.356890459363958e-05,
|
| 299 |
-
"loss": 0.9941,
|
| 300 |
"step": 220
|
| 301 |
},
|
| 302 |
{
|
| 303 |
-
"epoch":
|
| 304 |
-
"
|
| 305 |
-
"
|
| 306 |
-
"
|
|
|
|
|
|
|
|
|
|
| 307 |
},
|
| 308 |
{
|
| 309 |
-
"epoch":
|
| 310 |
-
"learning_rate":
|
| 311 |
-
"loss": 0.
|
| 312 |
"step": 230
|
| 313 |
},
|
| 314 |
{
|
| 315 |
-
"epoch":
|
| 316 |
-
"learning_rate":
|
| 317 |
-
"loss":
|
| 318 |
-
"step": 235
|
| 319 |
-
},
|
| 320 |
-
{
|
| 321 |
-
"epoch": 3.76,
|
| 322 |
-
"learning_rate": 2.6501766784452298e-05,
|
| 323 |
-
"loss": 0.9585,
|
| 324 |
"step": 240
|
| 325 |
},
|
| 326 |
{
|
| 327 |
-
"epoch":
|
| 328 |
-
"learning_rate":
|
| 329 |
-
"loss": 0.
|
| 330 |
-
"step": 245
|
| 331 |
-
},
|
| 332 |
-
{
|
| 333 |
-
"epoch": 3.92,
|
| 334 |
-
"learning_rate": 2.296819787985866e-05,
|
| 335 |
-
"loss": 0.9441,
|
| 336 |
"step": 250
|
| 337 |
},
|
| 338 |
{
|
| 339 |
-
"epoch":
|
| 340 |
-
"
|
| 341 |
-
"
|
| 342 |
-
"
|
| 343 |
-
|
| 344 |
-
|
| 345 |
-
"
|
| 346 |
-
"eval_accuracy": 0.7318627450980392,
|
| 347 |
-
"eval_loss": 0.8730877041816711,
|
| 348 |
-
"eval_runtime": 38.4534,
|
| 349 |
-
"eval_samples_per_second": 53.051,
|
| 350 |
-
"eval_steps_per_second": 1.664,
|
| 351 |
-
"step": 255
|
| 352 |
-
},
|
| 353 |
-
{
|
| 354 |
-
"epoch": 4.08,
|
| 355 |
-
"learning_rate": 1.9434628975265016e-05,
|
| 356 |
-
"loss": 0.9134,
|
| 357 |
-
"step": 260
|
| 358 |
-
},
|
| 359 |
-
{
|
| 360 |
-
"epoch": 4.16,
|
| 361 |
-
"learning_rate": 1.76678445229682e-05,
|
| 362 |
-
"loss": 0.8873,
|
| 363 |
-
"step": 265
|
| 364 |
-
},
|
| 365 |
-
{
|
| 366 |
-
"epoch": 4.24,
|
| 367 |
-
"learning_rate": 1.5901060070671377e-05,
|
| 368 |
-
"loss": 0.898,
|
| 369 |
-
"step": 270
|
| 370 |
-
},
|
| 371 |
-
{
|
| 372 |
-
"epoch": 4.31,
|
| 373 |
-
"learning_rate": 1.413427561837456e-05,
|
| 374 |
-
"loss": 0.8586,
|
| 375 |
-
"step": 275
|
| 376 |
-
},
|
| 377 |
-
{
|
| 378 |
-
"epoch": 4.39,
|
| 379 |
-
"learning_rate": 1.236749116607774e-05,
|
| 380 |
-
"loss": 0.8358,
|
| 381 |
-
"step": 280
|
| 382 |
-
},
|
| 383 |
-
{
|
| 384 |
-
"epoch": 4.47,
|
| 385 |
-
"learning_rate": 1.060070671378092e-05,
|
| 386 |
-
"loss": 0.9083,
|
| 387 |
-
"step": 285
|
| 388 |
-
},
|
| 389 |
-
{
|
| 390 |
-
"epoch": 4.55,
|
| 391 |
-
"learning_rate": 8.8339222614841e-06,
|
| 392 |
-
"loss": 0.9129,
|
| 393 |
-
"step": 290
|
| 394 |
-
},
|
| 395 |
-
{
|
| 396 |
-
"epoch": 4.63,
|
| 397 |
-
"learning_rate": 7.06713780918728e-06,
|
| 398 |
-
"loss": 0.8401,
|
| 399 |
-
"step": 295
|
| 400 |
-
},
|
| 401 |
-
{
|
| 402 |
-
"epoch": 4.71,
|
| 403 |
-
"learning_rate": 5.30035335689046e-06,
|
| 404 |
-
"loss": 0.9357,
|
| 405 |
-
"step": 300
|
| 406 |
-
},
|
| 407 |
-
{
|
| 408 |
-
"epoch": 4.78,
|
| 409 |
-
"learning_rate": 3.53356890459364e-06,
|
| 410 |
-
"loss": 0.82,
|
| 411 |
-
"step": 305
|
| 412 |
-
},
|
| 413 |
-
{
|
| 414 |
-
"epoch": 4.86,
|
| 415 |
-
"learning_rate": 1.76678445229682e-06,
|
| 416 |
-
"loss": 0.9039,
|
| 417 |
-
"step": 310
|
| 418 |
},
|
| 419 |
{
|
| 420 |
-
"epoch":
|
| 421 |
-
"
|
| 422 |
-
"
|
| 423 |
-
"
|
| 424 |
-
|
| 425 |
-
|
| 426 |
-
"
|
| 427 |
-
"eval_accuracy": 0.7436274509803922,
|
| 428 |
-
"eval_loss": 0.831763505935669,
|
| 429 |
-
"eval_runtime": 39.256,
|
| 430 |
-
"eval_samples_per_second": 51.967,
|
| 431 |
-
"eval_steps_per_second": 1.63,
|
| 432 |
-
"step": 315
|
| 433 |
-
},
|
| 434 |
-
{
|
| 435 |
-
"epoch": 4.94,
|
| 436 |
-
"step": 315,
|
| 437 |
-
"total_flos": 1.5485654902849044e+18,
|
| 438 |
-
"train_loss": 1.3921774001348586,
|
| 439 |
-
"train_runtime": 1093.5355,
|
| 440 |
-
"train_samples_per_second": 37.306,
|
| 441 |
-
"train_steps_per_second": 0.288
|
| 442 |
}
|
| 443 |
],
|
| 444 |
-
"logging_steps":
|
| 445 |
-
"max_steps":
|
| 446 |
-
"num_train_epochs":
|
| 447 |
"save_steps": 500,
|
| 448 |
-
"total_flos":
|
| 449 |
"trial_name": null,
|
| 450 |
"trial_params": null
|
| 451 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.7700980392156863,
|
| 3 |
+
"best_model_checkpoint": "resnet-101-finetuned-CivilEng11k/checkpoint-229",
|
| 4 |
+
"epoch": 9.803921568627452,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 250,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.39,
|
| 13 |
+
"learning_rate": 4e-05,
|
| 14 |
+
"loss": 2.6157,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 15 |
"step": 10
|
| 16 |
},
|
| 17 |
{
|
| 18 |
+
"epoch": 0.78,
|
| 19 |
+
"learning_rate": 8e-05,
|
| 20 |
+
"loss": 2.5769,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 21 |
"step": 20
|
| 22 |
},
|
| 23 |
{
|
| 24 |
+
"epoch": 0.98,
|
| 25 |
+
"eval_accuracy": 0.5269607843137255,
|
| 26 |
+
"eval_loss": 2.49281907081604,
|
| 27 |
+
"eval_runtime": 40.5904,
|
| 28 |
+
"eval_samples_per_second": 50.258,
|
| 29 |
+
"eval_steps_per_second": 1.577,
|
| 30 |
"step": 25
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 1.18,
|
| 34 |
+
"learning_rate": 9.777777777777778e-05,
|
| 35 |
+
"loss": 2.4933,
|
| 36 |
"step": 30
|
| 37 |
},
|
| 38 |
{
|
| 39 |
+
"epoch": 1.57,
|
| 40 |
+
"learning_rate": 9.333333333333334e-05,
|
| 41 |
+
"loss": 2.3726,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 42 |
"step": 40
|
| 43 |
},
|
| 44 |
{
|
| 45 |
+
"epoch": 1.96,
|
| 46 |
+
"learning_rate": 8.888888888888889e-05,
|
| 47 |
+
"loss": 2.2271,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 48 |
"step": 50
|
| 49 |
},
|
| 50 |
{
|
| 51 |
+
"epoch": 2.0,
|
| 52 |
+
"eval_accuracy": 0.5284313725490196,
|
| 53 |
+
"eval_loss": 2.184375047683716,
|
| 54 |
+
"eval_runtime": 41.3607,
|
| 55 |
+
"eval_samples_per_second": 49.322,
|
| 56 |
+
"eval_steps_per_second": 1.547,
|
| 57 |
+
"step": 51
|
| 58 |
},
|
| 59 |
{
|
| 60 |
+
"epoch": 2.35,
|
| 61 |
+
"learning_rate": 8.444444444444444e-05,
|
| 62 |
+
"loss": 2.004,
|
| 63 |
"step": 60
|
| 64 |
},
|
| 65 |
{
|
| 66 |
+
"epoch": 2.75,
|
| 67 |
+
"learning_rate": 8e-05,
|
| 68 |
+
"loss": 1.6261,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 69 |
"step": 70
|
| 70 |
},
|
| 71 |
{
|
| 72 |
+
"epoch": 2.98,
|
| 73 |
+
"eval_accuracy": 0.5269607843137255,
|
| 74 |
+
"eval_loss": 1.4098289012908936,
|
| 75 |
+
"eval_runtime": 40.6734,
|
| 76 |
+
"eval_samples_per_second": 50.156,
|
| 77 |
+
"eval_steps_per_second": 1.574,
|
| 78 |
+
"step": 76
|
| 79 |
},
|
| 80 |
{
|
| 81 |
+
"epoch": 3.14,
|
| 82 |
+
"learning_rate": 7.555555555555556e-05,
|
| 83 |
+
"loss": 1.4818,
|
| 84 |
"step": 80
|
| 85 |
},
|
| 86 |
{
|
| 87 |
+
"epoch": 3.53,
|
| 88 |
+
"learning_rate": 7.111111111111112e-05,
|
| 89 |
+
"loss": 1.3531,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 90 |
"step": 90
|
| 91 |
},
|
| 92 |
{
|
| 93 |
+
"epoch": 3.92,
|
| 94 |
+
"learning_rate": 6.666666666666667e-05,
|
| 95 |
+
"loss": 1.2715,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 96 |
"step": 100
|
| 97 |
},
|
| 98 |
{
|
| 99 |
+
"epoch": 4.0,
|
| 100 |
+
"eval_accuracy": 0.5799019607843138,
|
| 101 |
+
"eval_loss": 1.204033613204956,
|
| 102 |
+
"eval_runtime": 41.3797,
|
| 103 |
+
"eval_samples_per_second": 49.3,
|
| 104 |
+
"eval_steps_per_second": 1.547,
|
| 105 |
+
"step": 102
|
| 106 |
},
|
| 107 |
{
|
| 108 |
+
"epoch": 4.31,
|
| 109 |
+
"learning_rate": 6.222222222222222e-05,
|
| 110 |
+
"loss": 1.1843,
|
| 111 |
"step": 110
|
| 112 |
},
|
| 113 |
{
|
| 114 |
+
"epoch": 4.71,
|
| 115 |
+
"learning_rate": 5.7777777777777776e-05,
|
| 116 |
+
"loss": 1.1368,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 117 |
"step": 120
|
| 118 |
},
|
| 119 |
{
|
| 120 |
+
"epoch": 4.98,
|
| 121 |
+
"eval_accuracy": 0.6852941176470588,
|
| 122 |
+
"eval_loss": 1.0043939352035522,
|
| 123 |
+
"eval_runtime": 40.9737,
|
| 124 |
+
"eval_samples_per_second": 49.788,
|
| 125 |
+
"eval_steps_per_second": 1.562,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 126 |
"step": 127
|
| 127 |
},
|
| 128 |
{
|
| 129 |
+
"epoch": 5.1,
|
| 130 |
+
"learning_rate": 5.333333333333333e-05,
|
| 131 |
+
"loss": 1.0742,
|
| 132 |
"step": 130
|
| 133 |
},
|
| 134 |
{
|
| 135 |
+
"epoch": 5.49,
|
| 136 |
+
"learning_rate": 4.888888888888889e-05,
|
| 137 |
+
"loss": 0.9825,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 138 |
"step": 140
|
| 139 |
},
|
| 140 |
{
|
| 141 |
+
"epoch": 5.88,
|
| 142 |
+
"learning_rate": 4.4444444444444447e-05,
|
| 143 |
+
"loss": 0.9366,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 144 |
"step": 150
|
| 145 |
},
|
| 146 |
{
|
| 147 |
+
"epoch": 6.0,
|
| 148 |
+
"eval_accuracy": 0.7455882352941177,
|
| 149 |
+
"eval_loss": 0.8462627530097961,
|
| 150 |
+
"eval_runtime": 40.2496,
|
| 151 |
+
"eval_samples_per_second": 50.684,
|
| 152 |
+
"eval_steps_per_second": 1.59,
|
| 153 |
+
"step": 153
|
| 154 |
},
|
| 155 |
{
|
| 156 |
+
"epoch": 6.27,
|
| 157 |
+
"learning_rate": 4e-05,
|
| 158 |
+
"loss": 0.899,
|
| 159 |
"step": 160
|
| 160 |
},
|
| 161 |
{
|
| 162 |
+
"epoch": 6.67,
|
| 163 |
+
"learning_rate": 3.555555555555556e-05,
|
| 164 |
+
"loss": 0.8249,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 165 |
"step": 170
|
| 166 |
},
|
| 167 |
{
|
| 168 |
+
"epoch": 6.98,
|
| 169 |
+
"eval_accuracy": 0.7686274509803922,
|
| 170 |
+
"eval_loss": 0.7511501908302307,
|
| 171 |
+
"eval_runtime": 47.4571,
|
| 172 |
+
"eval_samples_per_second": 42.986,
|
| 173 |
+
"eval_steps_per_second": 1.349,
|
| 174 |
+
"step": 178
|
| 175 |
},
|
| 176 |
{
|
| 177 |
+
"epoch": 7.06,
|
| 178 |
+
"learning_rate": 3.111111111111111e-05,
|
| 179 |
+
"loss": 0.8197,
|
| 180 |
"step": 180
|
| 181 |
},
|
| 182 |
{
|
| 183 |
+
"epoch": 7.45,
|
| 184 |
+
"learning_rate": 2.6666666666666667e-05,
|
| 185 |
+
"loss": 0.7562,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 186 |
"step": 190
|
| 187 |
},
|
| 188 |
{
|
| 189 |
+
"epoch": 7.84,
|
| 190 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 191 |
+
"loss": 0.7635,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 192 |
"step": 200
|
| 193 |
},
|
| 194 |
{
|
| 195 |
+
"epoch": 8.0,
|
| 196 |
+
"eval_accuracy": 0.7612745098039215,
|
| 197 |
+
"eval_loss": 0.7078944444656372,
|
| 198 |
+
"eval_runtime": 40.9903,
|
| 199 |
+
"eval_samples_per_second": 49.768,
|
| 200 |
+
"eval_steps_per_second": 1.561,
|
| 201 |
+
"step": 204
|
| 202 |
},
|
| 203 |
{
|
| 204 |
+
"epoch": 8.24,
|
| 205 |
+
"learning_rate": 1.777777777777778e-05,
|
| 206 |
+
"loss": 0.7415,
|
| 207 |
"step": 210
|
| 208 |
},
|
| 209 |
{
|
| 210 |
+
"epoch": 8.63,
|
| 211 |
+
"learning_rate": 1.3333333333333333e-05,
|
| 212 |
+
"loss": 0.7213,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 213 |
"step": 220
|
| 214 |
},
|
| 215 |
{
|
| 216 |
+
"epoch": 8.98,
|
| 217 |
+
"eval_accuracy": 0.7700980392156863,
|
| 218 |
+
"eval_loss": 0.6762834191322327,
|
| 219 |
+
"eval_runtime": 40.8595,
|
| 220 |
+
"eval_samples_per_second": 49.927,
|
| 221 |
+
"eval_steps_per_second": 1.566,
|
| 222 |
+
"step": 229
|
| 223 |
},
|
| 224 |
{
|
| 225 |
+
"epoch": 9.02,
|
| 226 |
+
"learning_rate": 8.88888888888889e-06,
|
| 227 |
+
"loss": 0.6951,
|
| 228 |
"step": 230
|
| 229 |
},
|
| 230 |
{
|
| 231 |
+
"epoch": 9.41,
|
| 232 |
+
"learning_rate": 4.444444444444445e-06,
|
| 233 |
+
"loss": 0.7295,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 234 |
"step": 240
|
| 235 |
},
|
| 236 |
{
|
| 237 |
+
"epoch": 9.8,
|
| 238 |
+
"learning_rate": 0.0,
|
| 239 |
+
"loss": 0.6905,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 240 |
"step": 250
|
| 241 |
},
|
| 242 |
{
|
| 243 |
+
"epoch": 9.8,
|
| 244 |
+
"eval_accuracy": 0.7676470588235295,
|
| 245 |
+
"eval_loss": 0.6683045625686646,
|
| 246 |
+
"eval_runtime": 39.5198,
|
| 247 |
+
"eval_samples_per_second": 51.62,
|
| 248 |
+
"eval_steps_per_second": 1.619,
|
| 249 |
+
"step": 250
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 250 |
},
|
| 251 |
{
|
| 252 |
+
"epoch": 9.8,
|
| 253 |
+
"step": 250,
|
| 254 |
+
"total_flos": 3.072509726495183e+18,
|
| 255 |
+
"train_loss": 1.3191132202148437,
|
| 256 |
+
"train_runtime": 2261.7516,
|
| 257 |
+
"train_samples_per_second": 36.074,
|
| 258 |
+
"train_steps_per_second": 0.111
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 259 |
}
|
| 260 |
],
|
| 261 |
+
"logging_steps": 10,
|
| 262 |
+
"max_steps": 250,
|
| 263 |
+
"num_train_epochs": 10,
|
| 264 |
"save_steps": 500,
|
| 265 |
+
"total_flos": 3.072509726495183e+18,
|
| 266 |
"trial_name": null,
|
| 267 |
"trial_params": null
|
| 268 |
}
|