End of training
Browse files- README.md +3 -3
- all_results.json +11 -11
- eval_results.json +6 -6
- runs/Dec15_10-38-19_666f530b4c9c/events.out.tfevents.1702638161.666f530b4c9c.589.1 +3 -0
- train_results.json +6 -6
- trainer_state.json +520 -544
README.md
CHANGED
|
@@ -22,7 +22,7 @@ model-index:
|
|
| 22 |
metrics:
|
| 23 |
- name: Accuracy
|
| 24 |
type: accuracy
|
| 25 |
-
value: 0.
|
| 26 |
---
|
| 27 |
|
| 28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -32,8 +32,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 32 |
|
| 33 |
This model is a fine-tuned version of [microsoft/swin-tiny-patch4-window7-224](https://huggingface.co/microsoft/swin-tiny-patch4-window7-224) on the imagefolder dataset.
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
-
- Loss: 0.
|
| 36 |
-
- Accuracy: 0.
|
| 37 |
|
| 38 |
## Model description
|
| 39 |
|
|
|
|
| 22 |
metrics:
|
| 23 |
- name: Accuracy
|
| 24 |
type: accuracy
|
| 25 |
+
value: 0.9777131782945736
|
| 26 |
---
|
| 27 |
|
| 28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 32 |
|
| 33 |
This model is a fine-tuned version of [microsoft/swin-tiny-patch4-window7-224](https://huggingface.co/microsoft/swin-tiny-patch4-window7-224) on the imagefolder dataset.
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
+
- Loss: 0.0748
|
| 36 |
+
- Accuracy: 0.9777
|
| 37 |
|
| 38 |
## Model description
|
| 39 |
|
all_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime": 6.
|
| 6 |
-
"eval_samples_per_second": 156.
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
-
"total_flos": 2.
|
| 9 |
-
"train_loss": 0.
|
| 10 |
-
"train_runtime":
|
| 11 |
-
"train_samples_per_second":
|
| 12 |
-
"train_steps_per_second": 0.
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 9.9,
|
| 3 |
+
"eval_accuracy": 0.9777131782945736,
|
| 4 |
+
"eval_loss": 0.0747998058795929,
|
| 5 |
+
"eval_runtime": 6.6002,
|
| 6 |
+
"eval_samples_per_second": 156.358,
|
| 7 |
+
"eval_steps_per_second": 5.0,
|
| 8 |
+
"total_flos": 2.2837931240094351e+18,
|
| 9 |
+
"train_loss": 0.11106966810507907,
|
| 10 |
+
"train_runtime": 1435.0993,
|
| 11 |
+
"train_samples_per_second": 64.671,
|
| 12 |
+
"train_steps_per_second": 0.502
|
| 13 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime": 6.
|
| 6 |
-
"eval_samples_per_second": 156.
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 9.9,
|
| 3 |
+
"eval_accuracy": 0.9777131782945736,
|
| 4 |
+
"eval_loss": 0.0747998058795929,
|
| 5 |
+
"eval_runtime": 6.6002,
|
| 6 |
+
"eval_samples_per_second": 156.358,
|
| 7 |
+
"eval_steps_per_second": 5.0
|
| 8 |
}
|
runs/Dec15_10-38-19_666f530b4c9c/events.out.tfevents.1702638161.666f530b4c9c.589.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:72c616f35c851598212f8f389c1c054fe6002301dd4540336697c18044db7c89
|
| 3 |
+
size 411
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos": 2.
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 9.9,
|
| 3 |
+
"total_flos": 2.2837931240094351e+18,
|
| 4 |
+
"train_loss": 0.11106966810507907,
|
| 5 |
+
"train_runtime": 1435.0993,
|
| 6 |
+
"train_samples_per_second": 64.671,
|
| 7 |
+
"train_steps_per_second": 0.502
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,1006 +1,982 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "swin-tiny-patch4-window7-224-blank_img/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": 0.07,
|
| 13 |
-
"learning_rate": 3.
|
| 14 |
-
"loss": 0.
|
| 15 |
"step": 5
|
| 16 |
},
|
| 17 |
{
|
| 18 |
"epoch": 0.14,
|
| 19 |
-
"learning_rate": 6.
|
| 20 |
-
"loss": 0.
|
| 21 |
"step": 10
|
| 22 |
},
|
| 23 |
{
|
| 24 |
-
"epoch": 0.
|
| 25 |
-
"learning_rate": 1.
|
| 26 |
-
"loss": 0.
|
| 27 |
"step": 15
|
| 28 |
},
|
| 29 |
{
|
| 30 |
"epoch": 0.27,
|
| 31 |
-
"learning_rate": 1.
|
| 32 |
-
"loss": 0.
|
| 33 |
"step": 20
|
| 34 |
},
|
| 35 |
{
|
| 36 |
"epoch": 0.34,
|
| 37 |
-
"learning_rate": 1.
|
| 38 |
-
"loss": 0.
|
| 39 |
"step": 25
|
| 40 |
},
|
| 41 |
{
|
| 42 |
"epoch": 0.41,
|
| 43 |
-
"learning_rate": 2.
|
| 44 |
-
"loss": 0.
|
| 45 |
"step": 30
|
| 46 |
},
|
| 47 |
{
|
| 48 |
-
"epoch": 0.
|
| 49 |
-
"learning_rate": 2.
|
| 50 |
-
"loss": 0.
|
| 51 |
"step": 35
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
-
"learning_rate": 2.
|
| 56 |
-
"loss": 0.
|
| 57 |
"step": 40
|
| 58 |
},
|
| 59 |
{
|
| 60 |
-
"epoch": 0.
|
| 61 |
-
"learning_rate": 3.
|
| 62 |
-
"loss": 0.
|
| 63 |
"step": 45
|
| 64 |
},
|
| 65 |
{
|
| 66 |
-
"epoch": 0.
|
| 67 |
-
"learning_rate": 3.
|
| 68 |
-
"loss": 0.
|
| 69 |
"step": 50
|
| 70 |
},
|
| 71 |
{
|
| 72 |
-
"epoch": 0.
|
| 73 |
-
"learning_rate": 3.
|
| 74 |
-
"loss": 0.
|
| 75 |
"step": 55
|
| 76 |
},
|
| 77 |
{
|
| 78 |
-
"epoch": 0.
|
| 79 |
-
"learning_rate": 4.
|
| 80 |
-
"loss": 0.
|
| 81 |
"step": 60
|
| 82 |
},
|
| 83 |
{
|
| 84 |
-
"epoch": 0.
|
| 85 |
-
"learning_rate": 4.
|
| 86 |
-
"loss": 0.
|
| 87 |
"step": 65
|
| 88 |
},
|
| 89 |
{
|
| 90 |
-
"epoch": 0.
|
| 91 |
-
"learning_rate": 4.
|
| 92 |
-
"loss": 0.
|
| 93 |
"step": 70
|
| 94 |
},
|
| 95 |
{
|
| 96 |
-
"epoch":
|
| 97 |
-
"eval_accuracy": 0.
|
| 98 |
-
"eval_loss": 0.
|
| 99 |
-
"eval_runtime":
|
| 100 |
-
"eval_samples_per_second":
|
| 101 |
-
"eval_steps_per_second": 4.
|
| 102 |
-
"step":
|
| 103 |
},
|
| 104 |
{
|
| 105 |
-
"epoch": 1.
|
| 106 |
-
"learning_rate": 4.
|
| 107 |
-
"loss": 0.
|
| 108 |
"step": 75
|
| 109 |
},
|
| 110 |
{
|
| 111 |
-
"epoch": 1.
|
| 112 |
-
"learning_rate": 4.
|
| 113 |
-
"loss": 0.
|
| 114 |
"step": 80
|
| 115 |
},
|
| 116 |
{
|
| 117 |
-
"epoch": 1.
|
| 118 |
-
"learning_rate": 4.
|
| 119 |
-
"loss": 0.
|
| 120 |
"step": 85
|
| 121 |
},
|
| 122 |
{
|
| 123 |
-
"epoch": 1.
|
| 124 |
-
"learning_rate": 4.
|
| 125 |
-
"loss": 0.
|
| 126 |
"step": 90
|
| 127 |
},
|
| 128 |
{
|
| 129 |
-
"epoch": 1.
|
| 130 |
-
"learning_rate": 4.
|
| 131 |
-
"loss": 0.
|
| 132 |
"step": 95
|
| 133 |
},
|
| 134 |
{
|
| 135 |
-
"epoch": 1.
|
| 136 |
-
"learning_rate": 4.
|
| 137 |
-
"loss": 0.
|
| 138 |
"step": 100
|
| 139 |
},
|
| 140 |
{
|
| 141 |
-
"epoch": 1.
|
| 142 |
-
"learning_rate": 4.
|
| 143 |
-
"loss": 0.
|
| 144 |
"step": 105
|
| 145 |
},
|
| 146 |
{
|
| 147 |
-
"epoch": 1.
|
| 148 |
-
"learning_rate": 4.
|
| 149 |
-
"loss": 0.
|
| 150 |
"step": 110
|
| 151 |
},
|
| 152 |
{
|
| 153 |
-
"epoch": 1.
|
| 154 |
-
"learning_rate": 4.
|
| 155 |
-
"loss": 0.
|
| 156 |
"step": 115
|
| 157 |
},
|
| 158 |
{
|
| 159 |
-
"epoch": 1.
|
| 160 |
-
"learning_rate": 4.
|
| 161 |
-
"loss": 0.
|
| 162 |
"step": 120
|
| 163 |
},
|
| 164 |
{
|
| 165 |
-
"epoch": 1.
|
| 166 |
-
"learning_rate": 4.
|
| 167 |
-
"loss": 0.
|
| 168 |
"step": 125
|
| 169 |
},
|
| 170 |
{
|
| 171 |
-
"epoch": 1.
|
| 172 |
-
"learning_rate": 4.
|
| 173 |
-
"loss": 0.
|
| 174 |
"step": 130
|
| 175 |
},
|
| 176 |
{
|
| 177 |
-
"epoch": 1.
|
| 178 |
-
"learning_rate": 4.
|
| 179 |
-
"loss": 0.
|
| 180 |
"step": 135
|
| 181 |
},
|
| 182 |
{
|
| 183 |
-
"epoch": 1.
|
| 184 |
-
"learning_rate": 4.
|
| 185 |
-
"loss": 0.
|
| 186 |
"step": 140
|
| 187 |
},
|
| 188 |
{
|
| 189 |
-
"epoch": 1.
|
| 190 |
-
"learning_rate": 4.
|
| 191 |
-
"loss": 0.
|
| 192 |
"step": 145
|
| 193 |
},
|
| 194 |
{
|
| 195 |
-
"epoch":
|
| 196 |
-
"eval_accuracy": 0.
|
| 197 |
-
"eval_loss": 0.
|
| 198 |
-
"eval_runtime": 7.
|
| 199 |
-
"eval_samples_per_second":
|
| 200 |
-
"eval_steps_per_second": 4.
|
| 201 |
-
"step":
|
| 202 |
},
|
| 203 |
{
|
| 204 |
-
"epoch": 2.
|
| 205 |
-
"learning_rate": 4.
|
| 206 |
-
"loss": 0.
|
| 207 |
"step": 150
|
| 208 |
},
|
| 209 |
{
|
| 210 |
-
"epoch": 2.
|
| 211 |
-
"learning_rate": 4.
|
| 212 |
-
"loss": 0.
|
| 213 |
"step": 155
|
| 214 |
},
|
| 215 |
{
|
| 216 |
-
"epoch": 2.
|
| 217 |
-
"learning_rate": 4.
|
| 218 |
-
"loss": 0.
|
| 219 |
"step": 160
|
| 220 |
},
|
| 221 |
{
|
| 222 |
-
"epoch": 2.
|
| 223 |
-
"learning_rate": 4.
|
| 224 |
-
"loss": 0.
|
| 225 |
"step": 165
|
| 226 |
},
|
| 227 |
{
|
| 228 |
-
"epoch": 2.
|
| 229 |
-
"learning_rate": 4.
|
| 230 |
-
"loss": 0.
|
| 231 |
"step": 170
|
| 232 |
},
|
| 233 |
{
|
| 234 |
-
"epoch": 2.
|
| 235 |
-
"learning_rate": 4.
|
| 236 |
-
"loss": 0.
|
| 237 |
"step": 175
|
| 238 |
},
|
| 239 |
{
|
| 240 |
-
"epoch": 2.
|
| 241 |
-
"learning_rate": 4.
|
| 242 |
-
"loss": 0.
|
| 243 |
"step": 180
|
| 244 |
},
|
| 245 |
{
|
| 246 |
-
"epoch": 2.
|
| 247 |
-
"learning_rate": 4.
|
| 248 |
-
"loss": 0.
|
| 249 |
"step": 185
|
| 250 |
},
|
| 251 |
{
|
| 252 |
-
"epoch": 2.
|
| 253 |
-
"learning_rate": 4.
|
| 254 |
-
"loss": 0.
|
| 255 |
"step": 190
|
| 256 |
},
|
| 257 |
{
|
| 258 |
-
"epoch": 2.
|
| 259 |
-
"learning_rate": 4.
|
| 260 |
-
"loss": 0.
|
| 261 |
"step": 195
|
| 262 |
},
|
| 263 |
{
|
| 264 |
-
"epoch": 2.
|
| 265 |
-
"learning_rate": 4.
|
| 266 |
-
"loss": 0.
|
| 267 |
"step": 200
|
| 268 |
},
|
| 269 |
{
|
| 270 |
-
"epoch": 2.
|
| 271 |
-
"learning_rate":
|
| 272 |
-
"loss": 0.
|
| 273 |
"step": 205
|
| 274 |
},
|
| 275 |
{
|
| 276 |
-
"epoch": 2.
|
| 277 |
-
"learning_rate": 3.
|
| 278 |
-
"loss": 0.
|
| 279 |
"step": 210
|
| 280 |
},
|
| 281 |
{
|
| 282 |
-
"epoch": 2.
|
| 283 |
-
"learning_rate": 3.
|
| 284 |
-
"loss": 0.
|
| 285 |
"step": 215
|
| 286 |
},
|
| 287 |
{
|
| 288 |
-
"epoch":
|
| 289 |
-
"
|
| 290 |
-
"
|
| 291 |
-
"
|
|
|
|
|
|
|
|
|
|
| 292 |
},
|
| 293 |
{
|
| 294 |
-
"epoch": 3.
|
| 295 |
-
"
|
| 296 |
-
"
|
| 297 |
-
"
|
| 298 |
-
"eval_samples_per_second": 140.747,
|
| 299 |
-
"eval_steps_per_second": 4.415,
|
| 300 |
-
"step": 222
|
| 301 |
},
|
| 302 |
{
|
| 303 |
-
"epoch": 3.
|
| 304 |
-
"learning_rate": 3.
|
| 305 |
-
"loss": 0.
|
| 306 |
"step": 225
|
| 307 |
},
|
| 308 |
{
|
| 309 |
-
"epoch": 3.
|
| 310 |
-
"learning_rate": 3.
|
| 311 |
-
"loss": 0.
|
| 312 |
"step": 230
|
| 313 |
},
|
| 314 |
{
|
| 315 |
-
"epoch": 3.
|
| 316 |
-
"learning_rate": 3.
|
| 317 |
-
"loss": 0.
|
| 318 |
"step": 235
|
| 319 |
},
|
| 320 |
{
|
| 321 |
-
"epoch": 3.
|
| 322 |
-
"learning_rate": 3.
|
| 323 |
-
"loss": 0.
|
| 324 |
"step": 240
|
| 325 |
},
|
| 326 |
{
|
| 327 |
-
"epoch": 3.
|
| 328 |
-
"learning_rate": 3.
|
| 329 |
-
"loss": 0.
|
| 330 |
"step": 245
|
| 331 |
},
|
| 332 |
{
|
| 333 |
-
"epoch": 3.
|
| 334 |
-
"learning_rate": 3.
|
| 335 |
-
"loss": 0.
|
| 336 |
"step": 250
|
| 337 |
},
|
| 338 |
{
|
| 339 |
-
"epoch": 3.
|
| 340 |
-
"learning_rate": 3.
|
| 341 |
-
"loss": 0.
|
| 342 |
"step": 255
|
| 343 |
},
|
| 344 |
{
|
| 345 |
-
"epoch": 3.
|
| 346 |
-
"learning_rate": 3.
|
| 347 |
-
"loss": 0.
|
| 348 |
"step": 260
|
| 349 |
},
|
| 350 |
{
|
| 351 |
-
"epoch": 3.
|
| 352 |
-
"learning_rate": 3.
|
| 353 |
-
"loss": 0.
|
| 354 |
"step": 265
|
| 355 |
},
|
| 356 |
{
|
| 357 |
-
"epoch": 3.
|
| 358 |
-
"learning_rate": 3.
|
| 359 |
-
"loss": 0.
|
| 360 |
"step": 270
|
| 361 |
},
|
| 362 |
-
{
|
| 363 |
-
"epoch": 3.72,
|
| 364 |
-
"learning_rate": 3.490990990990991e-05,
|
| 365 |
-
"loss": 0.1991,
|
| 366 |
-
"step": 275
|
| 367 |
-
},
|
| 368 |
{
|
| 369 |
"epoch": 3.78,
|
| 370 |
-
"learning_rate": 3.
|
| 371 |
-
"loss": 0.
|
| 372 |
-
"step":
|
| 373 |
},
|
| 374 |
{
|
| 375 |
"epoch": 3.85,
|
| 376 |
-
"learning_rate": 3.
|
| 377 |
-
"loss": 0.
|
| 378 |
-
"step":
|
| 379 |
},
|
| 380 |
{
|
| 381 |
"epoch": 3.92,
|
| 382 |
-
"learning_rate": 3.
|
| 383 |
-
"loss": 0.
|
| 384 |
-
"step":
|
| 385 |
},
|
| 386 |
{
|
| 387 |
"epoch": 3.99,
|
| 388 |
-
"learning_rate": 3.
|
| 389 |
-
"loss": 0.
|
| 390 |
-
"step":
|
| 391 |
},
|
| 392 |
{
|
| 393 |
"epoch": 4.0,
|
| 394 |
-
"eval_accuracy": 0.
|
| 395 |
-
"eval_loss": 0.
|
| 396 |
-
"eval_runtime":
|
| 397 |
-
"eval_samples_per_second":
|
| 398 |
-
"eval_steps_per_second":
|
| 399 |
-
"step":
|
| 400 |
},
|
| 401 |
{
|
| 402 |
"epoch": 4.05,
|
| 403 |
-
"learning_rate": 3.
|
| 404 |
-
"loss": 0.
|
| 405 |
-
"step":
|
| 406 |
},
|
| 407 |
{
|
| 408 |
"epoch": 4.12,
|
| 409 |
-
"learning_rate": 3.
|
| 410 |
-
"loss": 0.
|
| 411 |
-
"step":
|
| 412 |
},
|
| 413 |
{
|
| 414 |
"epoch": 4.19,
|
| 415 |
-
"learning_rate": 3.
|
| 416 |
-
"loss": 0.
|
| 417 |
-
"step":
|
| 418 |
},
|
| 419 |
{
|
| 420 |
"epoch": 4.26,
|
| 421 |
-
"learning_rate": 3.
|
| 422 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 423 |
"step": 315
|
| 424 |
},
|
| 425 |
{
|
| 426 |
-
"epoch": 4.
|
| 427 |
-
"learning_rate": 3.
|
| 428 |
-
"loss": 0.
|
| 429 |
"step": 320
|
| 430 |
},
|
| 431 |
{
|
| 432 |
-
"epoch": 4.
|
| 433 |
-
"learning_rate": 3.
|
| 434 |
-
"loss": 0.
|
| 435 |
"step": 325
|
| 436 |
},
|
| 437 |
{
|
| 438 |
-
"epoch": 4.
|
| 439 |
-
"learning_rate": 3.
|
| 440 |
-
"loss": 0.
|
| 441 |
"step": 330
|
| 442 |
},
|
| 443 |
{
|
| 444 |
-
"epoch": 4.
|
| 445 |
-
"learning_rate":
|
| 446 |
-
"loss": 0.
|
| 447 |
"step": 335
|
| 448 |
},
|
| 449 |
{
|
| 450 |
-
"epoch": 4.
|
| 451 |
-
"learning_rate":
|
| 452 |
-
"loss": 0.
|
| 453 |
"step": 340
|
| 454 |
},
|
| 455 |
{
|
| 456 |
-
"epoch": 4.
|
| 457 |
-
"learning_rate": 2.
|
| 458 |
-
"loss": 0.
|
| 459 |
"step": 345
|
| 460 |
},
|
| 461 |
{
|
| 462 |
-
"epoch": 4.
|
| 463 |
-
"learning_rate": 2.
|
| 464 |
-
"loss": 0.
|
| 465 |
"step": 350
|
| 466 |
},
|
| 467 |
{
|
| 468 |
-
"epoch": 4.
|
| 469 |
-
"learning_rate": 2.
|
| 470 |
-
"loss": 0.
|
| 471 |
"step": 355
|
| 472 |
},
|
| 473 |
{
|
| 474 |
-
"epoch": 4.
|
| 475 |
-
"learning_rate": 2.
|
| 476 |
-
"loss": 0.
|
| 477 |
"step": 360
|
| 478 |
},
|
| 479 |
{
|
| 480 |
-
"epoch": 4.
|
| 481 |
-
"
|
| 482 |
-
"
|
| 483 |
-
"
|
|
|
|
|
|
|
|
|
|
| 484 |
},
|
| 485 |
{
|
| 486 |
-
"epoch": 5.
|
| 487 |
-
"learning_rate": 2.
|
| 488 |
-
"loss": 0.
|
| 489 |
-
"step":
|
| 490 |
},
|
| 491 |
{
|
| 492 |
-
"epoch": 5.
|
| 493 |
-
"
|
| 494 |
-
"
|
| 495 |
-
"eval_runtime": 6.7978,
|
| 496 |
-
"eval_samples_per_second": 154.755,
|
| 497 |
-
"eval_steps_per_second": 4.854,
|
| 498 |
"step": 370
|
| 499 |
},
|
| 500 |
{
|
| 501 |
-
"epoch": 5.
|
| 502 |
-
"learning_rate": 2.
|
| 503 |
-
"loss": 0.
|
| 504 |
"step": 375
|
| 505 |
},
|
| 506 |
{
|
| 507 |
-
"epoch": 5.
|
| 508 |
-
"learning_rate": 2.
|
| 509 |
-
"loss": 0.
|
| 510 |
"step": 380
|
| 511 |
},
|
| 512 |
{
|
| 513 |
-
"epoch": 5.
|
| 514 |
-
"learning_rate": 2.
|
| 515 |
-
"loss": 0.
|
| 516 |
"step": 385
|
| 517 |
},
|
| 518 |
{
|
| 519 |
-
"epoch": 5.
|
| 520 |
-
"learning_rate": 2.
|
| 521 |
-
"loss": 0.
|
| 522 |
"step": 390
|
| 523 |
},
|
| 524 |
{
|
| 525 |
-
"epoch": 5.
|
| 526 |
-
"learning_rate": 2.
|
| 527 |
-
"loss": 0.
|
| 528 |
"step": 395
|
| 529 |
},
|
| 530 |
{
|
| 531 |
-
"epoch": 5.
|
| 532 |
-
"learning_rate": 2.
|
| 533 |
-
"loss": 0.
|
| 534 |
"step": 400
|
| 535 |
},
|
| 536 |
{
|
| 537 |
-
"epoch": 5.
|
| 538 |
-
"learning_rate": 2.
|
| 539 |
-
"loss": 0.
|
| 540 |
"step": 405
|
| 541 |
},
|
| 542 |
{
|
| 543 |
-
"epoch": 5.
|
| 544 |
-
"learning_rate": 2.
|
| 545 |
-
"loss": 0.
|
| 546 |
"step": 410
|
| 547 |
},
|
| 548 |
{
|
| 549 |
-
"epoch": 5.
|
| 550 |
-
"learning_rate": 2.
|
| 551 |
-
"loss": 0.
|
| 552 |
"step": 415
|
| 553 |
},
|
| 554 |
{
|
| 555 |
-
"epoch": 5.
|
| 556 |
-
"learning_rate": 2.
|
| 557 |
-
"loss": 0.
|
| 558 |
"step": 420
|
| 559 |
},
|
| 560 |
{
|
| 561 |
-
"epoch": 5.
|
| 562 |
-
"learning_rate": 2.
|
| 563 |
-
"loss": 0.
|
| 564 |
"step": 425
|
| 565 |
},
|
| 566 |
{
|
| 567 |
-
"epoch": 5.
|
| 568 |
-
"learning_rate": 2.
|
| 569 |
-
"loss": 0.
|
| 570 |
"step": 430
|
| 571 |
},
|
| 572 |
{
|
| 573 |
-
"epoch": 5.
|
| 574 |
-
"learning_rate": 2.
|
| 575 |
-
"loss": 0.
|
| 576 |
"step": 435
|
| 577 |
},
|
| 578 |
{
|
| 579 |
-
"epoch": 5.
|
| 580 |
-
"
|
| 581 |
-
"
|
| 582 |
-
"
|
|
|
|
|
|
|
|
|
|
| 583 |
},
|
| 584 |
{
|
| 585 |
-
"epoch": 6.
|
| 586 |
-
"
|
| 587 |
-
"
|
| 588 |
-
"
|
| 589 |
-
"eval_samples_per_second": 156.609,
|
| 590 |
-
"eval_steps_per_second": 4.913,
|
| 591 |
-
"step": 444
|
| 592 |
},
|
| 593 |
{
|
| 594 |
-
"epoch": 6.
|
| 595 |
-
"learning_rate": 2.
|
| 596 |
-
"loss": 0.
|
| 597 |
"step": 445
|
| 598 |
},
|
| 599 |
{
|
| 600 |
-
"epoch": 6.
|
| 601 |
-
"learning_rate": 2.
|
| 602 |
-
"loss": 0.
|
| 603 |
"step": 450
|
| 604 |
},
|
| 605 |
{
|
| 606 |
-
"epoch": 6.
|
| 607 |
-
"learning_rate": 2.
|
| 608 |
-
"loss": 0.
|
| 609 |
"step": 455
|
| 610 |
},
|
| 611 |
{
|
| 612 |
-
"epoch": 6.
|
| 613 |
-
"learning_rate": 2.
|
| 614 |
-
"loss": 0.
|
| 615 |
"step": 460
|
| 616 |
},
|
| 617 |
{
|
| 618 |
-
"epoch": 6.
|
| 619 |
-
"learning_rate":
|
| 620 |
-
"loss": 0.
|
| 621 |
"step": 465
|
| 622 |
},
|
| 623 |
{
|
| 624 |
-
"epoch": 6.
|
| 625 |
-
"learning_rate":
|
| 626 |
-
"loss": 0.
|
| 627 |
"step": 470
|
| 628 |
},
|
| 629 |
{
|
| 630 |
-
"epoch": 6.
|
| 631 |
-
"learning_rate": 1.
|
| 632 |
-
"loss": 0.
|
| 633 |
"step": 475
|
| 634 |
},
|
| 635 |
{
|
| 636 |
-
"epoch": 6.
|
| 637 |
-
"learning_rate": 1.
|
| 638 |
-
"loss": 0.
|
| 639 |
"step": 480
|
| 640 |
},
|
| 641 |
{
|
| 642 |
-
"epoch": 6.
|
| 643 |
-
"learning_rate": 1.
|
| 644 |
-
"loss": 0.
|
| 645 |
"step": 485
|
| 646 |
},
|
| 647 |
{
|
| 648 |
-
"epoch": 6.
|
| 649 |
-
"learning_rate": 1.
|
| 650 |
-
"loss": 0.
|
| 651 |
"step": 490
|
| 652 |
},
|
| 653 |
{
|
| 654 |
-
"epoch": 6.
|
| 655 |
-
"learning_rate": 1.
|
| 656 |
-
"loss": 0.
|
| 657 |
"step": 495
|
| 658 |
},
|
| 659 |
{
|
| 660 |
-
"epoch": 6.
|
| 661 |
-
"learning_rate": 1.
|
| 662 |
-
"loss": 0.
|
| 663 |
"step": 500
|
| 664 |
},
|
| 665 |
{
|
| 666 |
-
"epoch": 6.
|
| 667 |
-
"learning_rate": 1.
|
| 668 |
-
"loss": 0.
|
| 669 |
"step": 505
|
| 670 |
},
|
| 671 |
{
|
| 672 |
-
"epoch":
|
| 673 |
-
"
|
| 674 |
-
"
|
| 675 |
-
"
|
|
|
|
|
|
|
|
|
|
| 676 |
},
|
| 677 |
{
|
| 678 |
-
"epoch":
|
| 679 |
-
"learning_rate": 1.
|
| 680 |
-
"loss": 0.
|
| 681 |
-
"step":
|
| 682 |
},
|
| 683 |
{
|
| 684 |
-
"epoch": 7.
|
| 685 |
-
"
|
| 686 |
-
"
|
| 687 |
-
"
|
| 688 |
-
"eval_samples_per_second": 143.339,
|
| 689 |
-
"eval_steps_per_second": 4.496,
|
| 690 |
-
"step": 518
|
| 691 |
},
|
| 692 |
{
|
| 693 |
-
"epoch": 7.
|
| 694 |
-
"learning_rate": 1.
|
| 695 |
-
"loss": 0.
|
| 696 |
"step": 520
|
| 697 |
},
|
| 698 |
{
|
| 699 |
-
"epoch": 7.
|
| 700 |
-
"learning_rate": 1.
|
| 701 |
-
"loss": 0.
|
| 702 |
"step": 525
|
| 703 |
},
|
| 704 |
{
|
| 705 |
-
"epoch": 7.
|
| 706 |
-
"learning_rate": 1.
|
| 707 |
-
"loss": 0.
|
| 708 |
"step": 530
|
| 709 |
},
|
| 710 |
{
|
| 711 |
-
"epoch": 7.
|
| 712 |
-
"learning_rate": 1.
|
| 713 |
-
"loss": 0.
|
| 714 |
"step": 535
|
| 715 |
},
|
| 716 |
{
|
| 717 |
-
"epoch": 7.
|
| 718 |
-
"learning_rate": 1.
|
| 719 |
-
"loss": 0.
|
| 720 |
"step": 540
|
| 721 |
},
|
| 722 |
{
|
| 723 |
-
"epoch": 7.
|
| 724 |
-
"learning_rate": 1.
|
| 725 |
-
"loss": 0.
|
| 726 |
"step": 545
|
| 727 |
},
|
| 728 |
{
|
| 729 |
-
"epoch": 7.
|
| 730 |
-
"learning_rate": 1.
|
| 731 |
-
"loss": 0.
|
| 732 |
"step": 550
|
| 733 |
},
|
| 734 |
{
|
| 735 |
-
"epoch": 7.
|
| 736 |
-
"learning_rate": 1.
|
| 737 |
-
"loss": 0.
|
| 738 |
"step": 555
|
| 739 |
},
|
| 740 |
{
|
| 741 |
-
"epoch": 7.
|
| 742 |
-
"learning_rate": 1.
|
| 743 |
-
"loss": 0.
|
| 744 |
"step": 560
|
| 745 |
},
|
| 746 |
{
|
| 747 |
-
"epoch": 7.
|
| 748 |
-
"learning_rate": 1.
|
| 749 |
-
"loss": 0.
|
| 750 |
"step": 565
|
| 751 |
},
|
| 752 |
{
|
| 753 |
-
"epoch": 7.
|
| 754 |
-
"learning_rate": 1.
|
| 755 |
-
"loss": 0.
|
| 756 |
"step": 570
|
| 757 |
},
|
| 758 |
{
|
| 759 |
-
"epoch": 7.
|
| 760 |
-
"learning_rate": 1.
|
| 761 |
-
"loss": 0.
|
| 762 |
"step": 575
|
| 763 |
},
|
| 764 |
{
|
| 765 |
-
"epoch": 7.
|
| 766 |
-
"learning_rate": 1.
|
| 767 |
-
"loss": 0.
|
| 768 |
"step": 580
|
| 769 |
},
|
| 770 |
{
|
| 771 |
-
"epoch":
|
| 772 |
-
"
|
| 773 |
-
"
|
| 774 |
-
"
|
|
|
|
|
|
|
|
|
|
| 775 |
},
|
| 776 |
{
|
| 777 |
-
"epoch":
|
| 778 |
-
"learning_rate": 1.
|
| 779 |
-
"loss": 0.
|
| 780 |
-
"step":
|
| 781 |
},
|
| 782 |
{
|
| 783 |
-
"epoch": 8.
|
| 784 |
-
"
|
| 785 |
-
"
|
| 786 |
-
"
|
| 787 |
-
"eval_samples_per_second": 139.861,
|
| 788 |
-
"eval_steps_per_second": 4.387,
|
| 789 |
-
"step": 592
|
| 790 |
},
|
| 791 |
{
|
| 792 |
-
"epoch": 8.
|
| 793 |
-
"learning_rate":
|
| 794 |
-
"loss": 0.
|
| 795 |
"step": 595
|
| 796 |
},
|
| 797 |
{
|
| 798 |
-
"epoch": 8.
|
| 799 |
-
"learning_rate":
|
| 800 |
-
"loss": 0.
|
| 801 |
"step": 600
|
| 802 |
},
|
| 803 |
{
|
| 804 |
-
"epoch": 8.
|
| 805 |
-
"learning_rate":
|
| 806 |
-
"loss": 0.
|
| 807 |
"step": 605
|
| 808 |
},
|
| 809 |
{
|
| 810 |
-
"epoch": 8.
|
| 811 |
-
"learning_rate":
|
| 812 |
-
"loss": 0.
|
| 813 |
"step": 610
|
| 814 |
},
|
| 815 |
{
|
| 816 |
-
"epoch": 8.
|
| 817 |
-
"learning_rate":
|
| 818 |
-
"loss": 0.
|
| 819 |
"step": 615
|
| 820 |
},
|
| 821 |
{
|
| 822 |
-
"epoch": 8.
|
| 823 |
-
"learning_rate":
|
| 824 |
-
"loss": 0.
|
| 825 |
"step": 620
|
| 826 |
},
|
| 827 |
{
|
| 828 |
-
"epoch": 8.
|
| 829 |
-
"learning_rate":
|
| 830 |
-
"loss": 0.
|
| 831 |
"step": 625
|
| 832 |
},
|
| 833 |
{
|
| 834 |
-
"epoch": 8.
|
| 835 |
-
"learning_rate":
|
| 836 |
-
"loss": 0.
|
| 837 |
"step": 630
|
| 838 |
},
|
| 839 |
{
|
| 840 |
-
"epoch": 8.
|
| 841 |
-
"learning_rate":
|
| 842 |
-
"loss": 0.
|
| 843 |
"step": 635
|
| 844 |
},
|
| 845 |
{
|
| 846 |
-
"epoch": 8.
|
| 847 |
-
"learning_rate":
|
| 848 |
-
"loss": 0.
|
| 849 |
"step": 640
|
| 850 |
},
|
| 851 |
{
|
| 852 |
-
"epoch": 8.
|
| 853 |
-
"learning_rate":
|
| 854 |
-
"loss": 0.
|
| 855 |
"step": 645
|
| 856 |
},
|
| 857 |
{
|
| 858 |
-
"epoch": 8.
|
| 859 |
-
"learning_rate":
|
| 860 |
-
"loss": 0.
|
| 861 |
"step": 650
|
| 862 |
},
|
| 863 |
{
|
| 864 |
-
"epoch": 8.
|
| 865 |
-
"
|
| 866 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 867 |
"step": 655
|
| 868 |
},
|
| 869 |
{
|
| 870 |
-
"epoch":
|
| 871 |
-
"learning_rate":
|
| 872 |
-
"loss": 0.
|
| 873 |
"step": 660
|
| 874 |
},
|
| 875 |
{
|
| 876 |
-
"epoch":
|
| 877 |
-
"learning_rate":
|
| 878 |
-
"loss": 0.
|
| 879 |
"step": 665
|
| 880 |
},
|
| 881 |
{
|
| 882 |
-
"epoch": 9.
|
| 883 |
-
"
|
| 884 |
-
"
|
| 885 |
-
"eval_runtime": 6.7103,
|
| 886 |
-
"eval_samples_per_second": 156.774,
|
| 887 |
-
"eval_steps_per_second": 4.918,
|
| 888 |
-
"step": 666
|
| 889 |
-
},
|
| 890 |
-
{
|
| 891 |
-
"epoch": 9.05,
|
| 892 |
-
"learning_rate": 5.255255255255255e-06,
|
| 893 |
-
"loss": 0.1613,
|
| 894 |
"step": 670
|
| 895 |
},
|
| 896 |
{
|
| 897 |
-
"epoch": 9.
|
| 898 |
-
"learning_rate":
|
| 899 |
-
"loss": 0.
|
| 900 |
"step": 675
|
| 901 |
},
|
| 902 |
{
|
| 903 |
-
"epoch": 9.
|
| 904 |
-
"learning_rate":
|
| 905 |
-
"loss": 0.
|
| 906 |
"step": 680
|
| 907 |
},
|
| 908 |
{
|
| 909 |
-
"epoch": 9.
|
| 910 |
-
"learning_rate":
|
| 911 |
-
"loss": 0.
|
| 912 |
"step": 685
|
| 913 |
},
|
| 914 |
{
|
| 915 |
-
"epoch": 9.
|
| 916 |
-
"learning_rate":
|
| 917 |
-
"loss": 0.
|
| 918 |
"step": 690
|
| 919 |
},
|
| 920 |
{
|
| 921 |
-
"epoch": 9.
|
| 922 |
-
"learning_rate":
|
| 923 |
-
"loss": 0.
|
| 924 |
"step": 695
|
| 925 |
},
|
| 926 |
{
|
| 927 |
-
"epoch": 9.
|
| 928 |
-
"learning_rate":
|
| 929 |
-
"loss": 0.
|
| 930 |
"step": 700
|
| 931 |
},
|
| 932 |
{
|
| 933 |
-
"epoch": 9.
|
| 934 |
-
"learning_rate":
|
| 935 |
-
"loss": 0.
|
| 936 |
"step": 705
|
| 937 |
},
|
| 938 |
{
|
| 939 |
-
"epoch": 9.
|
| 940 |
-
"learning_rate":
|
| 941 |
-
"loss": 0.
|
| 942 |
"step": 710
|
| 943 |
},
|
| 944 |
{
|
| 945 |
-
"epoch": 9.
|
| 946 |
-
"learning_rate":
|
| 947 |
-
"loss": 0.
|
| 948 |
"step": 715
|
| 949 |
},
|
| 950 |
{
|
| 951 |
-
"epoch": 9.
|
| 952 |
-
"learning_rate":
|
| 953 |
-
"loss": 0.
|
| 954 |
"step": 720
|
| 955 |
},
|
| 956 |
{
|
| 957 |
-
"epoch": 9.
|
| 958 |
-
"
|
| 959 |
-
"
|
| 960 |
-
"
|
| 961 |
-
|
| 962 |
-
|
| 963 |
-
"
|
| 964 |
-
"learning_rate": 7.507507507507508e-07,
|
| 965 |
-
"loss": 0.1692,
|
| 966 |
-
"step": 730
|
| 967 |
-
},
|
| 968 |
-
{
|
| 969 |
-
"epoch": 9.93,
|
| 970 |
-
"learning_rate": 3.753753753753754e-07,
|
| 971 |
-
"loss": 0.1998,
|
| 972 |
-
"step": 735
|
| 973 |
},
|
| 974 |
{
|
| 975 |
-
"epoch":
|
| 976 |
-
"
|
| 977 |
-
"
|
| 978 |
-
"
|
| 979 |
-
|
| 980 |
-
|
| 981 |
-
"
|
| 982 |
-
"eval_accuracy": 0.9420152091254753,
|
| 983 |
-
"eval_loss": 0.17660604417324066,
|
| 984 |
-
"eval_runtime": 6.7015,
|
| 985 |
-
"eval_samples_per_second": 156.979,
|
| 986 |
-
"eval_steps_per_second": 4.924,
|
| 987 |
-
"step": 740
|
| 988 |
-
},
|
| 989 |
-
{
|
| 990 |
-
"epoch": 10.0,
|
| 991 |
-
"step": 740,
|
| 992 |
-
"total_flos": 2.3513765580619776e+18,
|
| 993 |
-
"train_loss": 0.21299834186966354,
|
| 994 |
-
"train_runtime": 1451.9061,
|
| 995 |
-
"train_samples_per_second": 65.156,
|
| 996 |
-
"train_steps_per_second": 0.51
|
| 997 |
}
|
| 998 |
],
|
| 999 |
"logging_steps": 5,
|
| 1000 |
-
"max_steps":
|
| 1001 |
"num_train_epochs": 10,
|
| 1002 |
"save_steps": 500,
|
| 1003 |
-
"total_flos": 2.
|
| 1004 |
"trial_name": null,
|
| 1005 |
"trial_params": null
|
| 1006 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.9777131782945736,
|
| 3 |
+
"best_model_checkpoint": "swin-tiny-patch4-window7-224-blank_img/checkpoint-291",
|
| 4 |
+
"epoch": 9.896907216494846,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 720,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.07,
|
| 13 |
+
"learning_rate": 3.4722222222222224e-06,
|
| 14 |
+
"loss": 0.6582,
|
| 15 |
"step": 5
|
| 16 |
},
|
| 17 |
{
|
| 18 |
"epoch": 0.14,
|
| 19 |
+
"learning_rate": 6.944444444444445e-06,
|
| 20 |
+
"loss": 0.6212,
|
| 21 |
"step": 10
|
| 22 |
},
|
| 23 |
{
|
| 24 |
+
"epoch": 0.21,
|
| 25 |
+
"learning_rate": 1.0416666666666668e-05,
|
| 26 |
+
"loss": 0.513,
|
| 27 |
"step": 15
|
| 28 |
},
|
| 29 |
{
|
| 30 |
"epoch": 0.27,
|
| 31 |
+
"learning_rate": 1.388888888888889e-05,
|
| 32 |
+
"loss": 0.3706,
|
| 33 |
"step": 20
|
| 34 |
},
|
| 35 |
{
|
| 36 |
"epoch": 0.34,
|
| 37 |
+
"learning_rate": 1.736111111111111e-05,
|
| 38 |
+
"loss": 0.2605,
|
| 39 |
"step": 25
|
| 40 |
},
|
| 41 |
{
|
| 42 |
"epoch": 0.41,
|
| 43 |
+
"learning_rate": 2.0833333333333336e-05,
|
| 44 |
+
"loss": 0.2189,
|
| 45 |
"step": 30
|
| 46 |
},
|
| 47 |
{
|
| 48 |
+
"epoch": 0.48,
|
| 49 |
+
"learning_rate": 2.4305555555555558e-05,
|
| 50 |
+
"loss": 0.1749,
|
| 51 |
"step": 35
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.55,
|
| 55 |
+
"learning_rate": 2.777777777777778e-05,
|
| 56 |
+
"loss": 0.1763,
|
| 57 |
"step": 40
|
| 58 |
},
|
| 59 |
{
|
| 60 |
+
"epoch": 0.62,
|
| 61 |
+
"learning_rate": 3.125e-05,
|
| 62 |
+
"loss": 0.136,
|
| 63 |
"step": 45
|
| 64 |
},
|
| 65 |
{
|
| 66 |
+
"epoch": 0.69,
|
| 67 |
+
"learning_rate": 3.472222222222222e-05,
|
| 68 |
+
"loss": 0.1538,
|
| 69 |
"step": 50
|
| 70 |
},
|
| 71 |
{
|
| 72 |
+
"epoch": 0.76,
|
| 73 |
+
"learning_rate": 3.8194444444444444e-05,
|
| 74 |
+
"loss": 0.1284,
|
| 75 |
"step": 55
|
| 76 |
},
|
| 77 |
{
|
| 78 |
+
"epoch": 0.82,
|
| 79 |
+
"learning_rate": 4.166666666666667e-05,
|
| 80 |
+
"loss": 0.1346,
|
| 81 |
"step": 60
|
| 82 |
},
|
| 83 |
{
|
| 84 |
+
"epoch": 0.89,
|
| 85 |
+
"learning_rate": 4.5138888888888894e-05,
|
| 86 |
+
"loss": 0.1098,
|
| 87 |
"step": 65
|
| 88 |
},
|
| 89 |
{
|
| 90 |
+
"epoch": 0.96,
|
| 91 |
+
"learning_rate": 4.8611111111111115e-05,
|
| 92 |
+
"loss": 0.0898,
|
| 93 |
"step": 70
|
| 94 |
},
|
| 95 |
{
|
| 96 |
+
"epoch": 0.99,
|
| 97 |
+
"eval_accuracy": 0.9428294573643411,
|
| 98 |
+
"eval_loss": 0.1245269775390625,
|
| 99 |
+
"eval_runtime": 7.009,
|
| 100 |
+
"eval_samples_per_second": 147.239,
|
| 101 |
+
"eval_steps_per_second": 4.708,
|
| 102 |
+
"step": 72
|
| 103 |
},
|
| 104 |
{
|
| 105 |
+
"epoch": 1.03,
|
| 106 |
+
"learning_rate": 4.976851851851852e-05,
|
| 107 |
+
"loss": 0.1117,
|
| 108 |
"step": 75
|
| 109 |
},
|
| 110 |
{
|
| 111 |
+
"epoch": 1.1,
|
| 112 |
+
"learning_rate": 4.938271604938271e-05,
|
| 113 |
+
"loss": 0.0963,
|
| 114 |
"step": 80
|
| 115 |
},
|
| 116 |
{
|
| 117 |
+
"epoch": 1.17,
|
| 118 |
+
"learning_rate": 4.899691358024692e-05,
|
| 119 |
+
"loss": 0.1203,
|
| 120 |
"step": 85
|
| 121 |
},
|
| 122 |
{
|
| 123 |
+
"epoch": 1.24,
|
| 124 |
+
"learning_rate": 4.8611111111111115e-05,
|
| 125 |
+
"loss": 0.126,
|
| 126 |
"step": 90
|
| 127 |
},
|
| 128 |
{
|
| 129 |
+
"epoch": 1.31,
|
| 130 |
+
"learning_rate": 4.8225308641975306e-05,
|
| 131 |
+
"loss": 0.1428,
|
| 132 |
"step": 95
|
| 133 |
},
|
| 134 |
{
|
| 135 |
+
"epoch": 1.37,
|
| 136 |
+
"learning_rate": 4.783950617283951e-05,
|
| 137 |
+
"loss": 0.1215,
|
| 138 |
"step": 100
|
| 139 |
},
|
| 140 |
{
|
| 141 |
+
"epoch": 1.44,
|
| 142 |
+
"learning_rate": 4.745370370370371e-05,
|
| 143 |
+
"loss": 0.1363,
|
| 144 |
"step": 105
|
| 145 |
},
|
| 146 |
{
|
| 147 |
+
"epoch": 1.51,
|
| 148 |
+
"learning_rate": 4.70679012345679e-05,
|
| 149 |
+
"loss": 0.1432,
|
| 150 |
"step": 110
|
| 151 |
},
|
| 152 |
{
|
| 153 |
+
"epoch": 1.58,
|
| 154 |
+
"learning_rate": 4.66820987654321e-05,
|
| 155 |
+
"loss": 0.1444,
|
| 156 |
"step": 115
|
| 157 |
},
|
| 158 |
{
|
| 159 |
+
"epoch": 1.65,
|
| 160 |
+
"learning_rate": 4.62962962962963e-05,
|
| 161 |
+
"loss": 0.1221,
|
| 162 |
"step": 120
|
| 163 |
},
|
| 164 |
{
|
| 165 |
+
"epoch": 1.72,
|
| 166 |
+
"learning_rate": 4.591049382716049e-05,
|
| 167 |
+
"loss": 0.1159,
|
| 168 |
"step": 125
|
| 169 |
},
|
| 170 |
{
|
| 171 |
+
"epoch": 1.79,
|
| 172 |
+
"learning_rate": 4.5524691358024696e-05,
|
| 173 |
+
"loss": 0.1311,
|
| 174 |
"step": 130
|
| 175 |
},
|
| 176 |
{
|
| 177 |
+
"epoch": 1.86,
|
| 178 |
+
"learning_rate": 4.5138888888888894e-05,
|
| 179 |
+
"loss": 0.0965,
|
| 180 |
"step": 135
|
| 181 |
},
|
| 182 |
{
|
| 183 |
+
"epoch": 1.92,
|
| 184 |
+
"learning_rate": 4.4753086419753084e-05,
|
| 185 |
+
"loss": 0.1129,
|
| 186 |
"step": 140
|
| 187 |
},
|
| 188 |
{
|
| 189 |
+
"epoch": 1.99,
|
| 190 |
+
"learning_rate": 4.436728395061729e-05,
|
| 191 |
+
"loss": 0.152,
|
| 192 |
"step": 145
|
| 193 |
},
|
| 194 |
{
|
| 195 |
+
"epoch": 1.99,
|
| 196 |
+
"eval_accuracy": 0.9748062015503876,
|
| 197 |
+
"eval_loss": 0.08107755333185196,
|
| 198 |
+
"eval_runtime": 7.1051,
|
| 199 |
+
"eval_samples_per_second": 145.247,
|
| 200 |
+
"eval_steps_per_second": 4.645,
|
| 201 |
+
"step": 145
|
| 202 |
},
|
| 203 |
{
|
| 204 |
+
"epoch": 2.06,
|
| 205 |
+
"learning_rate": 4.3981481481481486e-05,
|
| 206 |
+
"loss": 0.1105,
|
| 207 |
"step": 150
|
| 208 |
},
|
| 209 |
{
|
| 210 |
+
"epoch": 2.13,
|
| 211 |
+
"learning_rate": 4.359567901234568e-05,
|
| 212 |
+
"loss": 0.0788,
|
| 213 |
"step": 155
|
| 214 |
},
|
| 215 |
{
|
| 216 |
+
"epoch": 2.2,
|
| 217 |
+
"learning_rate": 4.3209876543209875e-05,
|
| 218 |
+
"loss": 0.0988,
|
| 219 |
"step": 160
|
| 220 |
},
|
| 221 |
{
|
| 222 |
+
"epoch": 2.27,
|
| 223 |
+
"learning_rate": 4.282407407407408e-05,
|
| 224 |
+
"loss": 0.0963,
|
| 225 |
"step": 165
|
| 226 |
},
|
| 227 |
{
|
| 228 |
+
"epoch": 2.34,
|
| 229 |
+
"learning_rate": 4.243827160493827e-05,
|
| 230 |
+
"loss": 0.1201,
|
| 231 |
"step": 170
|
| 232 |
},
|
| 233 |
{
|
| 234 |
+
"epoch": 2.41,
|
| 235 |
+
"learning_rate": 4.205246913580247e-05,
|
| 236 |
+
"loss": 0.1001,
|
| 237 |
"step": 175
|
| 238 |
},
|
| 239 |
{
|
| 240 |
+
"epoch": 2.47,
|
| 241 |
+
"learning_rate": 4.166666666666667e-05,
|
| 242 |
+
"loss": 0.1118,
|
| 243 |
"step": 180
|
| 244 |
},
|
| 245 |
{
|
| 246 |
+
"epoch": 2.54,
|
| 247 |
+
"learning_rate": 4.128086419753087e-05,
|
| 248 |
+
"loss": 0.0948,
|
| 249 |
"step": 185
|
| 250 |
},
|
| 251 |
{
|
| 252 |
+
"epoch": 2.61,
|
| 253 |
+
"learning_rate": 4.089506172839506e-05,
|
| 254 |
+
"loss": 0.0941,
|
| 255 |
"step": 190
|
| 256 |
},
|
| 257 |
{
|
| 258 |
+
"epoch": 2.68,
|
| 259 |
+
"learning_rate": 4.0509259259259265e-05,
|
| 260 |
+
"loss": 0.1224,
|
| 261 |
"step": 195
|
| 262 |
},
|
| 263 |
{
|
| 264 |
+
"epoch": 2.75,
|
| 265 |
+
"learning_rate": 4.012345679012346e-05,
|
| 266 |
+
"loss": 0.0919,
|
| 267 |
"step": 200
|
| 268 |
},
|
| 269 |
{
|
| 270 |
+
"epoch": 2.82,
|
| 271 |
+
"learning_rate": 3.973765432098765e-05,
|
| 272 |
+
"loss": 0.1715,
|
| 273 |
"step": 205
|
| 274 |
},
|
| 275 |
{
|
| 276 |
+
"epoch": 2.89,
|
| 277 |
+
"learning_rate": 3.935185185185186e-05,
|
| 278 |
+
"loss": 0.1294,
|
| 279 |
"step": 210
|
| 280 |
},
|
| 281 |
{
|
| 282 |
+
"epoch": 2.96,
|
| 283 |
+
"learning_rate": 3.8966049382716055e-05,
|
| 284 |
+
"loss": 0.1235,
|
| 285 |
"step": 215
|
| 286 |
},
|
| 287 |
{
|
| 288 |
+
"epoch": 3.0,
|
| 289 |
+
"eval_accuracy": 0.9699612403100775,
|
| 290 |
+
"eval_loss": 0.09580326825380325,
|
| 291 |
+
"eval_runtime": 6.6469,
|
| 292 |
+
"eval_samples_per_second": 155.26,
|
| 293 |
+
"eval_steps_per_second": 4.965,
|
| 294 |
+
"step": 218
|
| 295 |
},
|
| 296 |
{
|
| 297 |
+
"epoch": 3.02,
|
| 298 |
+
"learning_rate": 3.8580246913580246e-05,
|
| 299 |
+
"loss": 0.0777,
|
| 300 |
+
"step": 220
|
|
|
|
|
|
|
|
|
|
| 301 |
},
|
| 302 |
{
|
| 303 |
+
"epoch": 3.09,
|
| 304 |
+
"learning_rate": 3.8194444444444444e-05,
|
| 305 |
+
"loss": 0.1301,
|
| 306 |
"step": 225
|
| 307 |
},
|
| 308 |
{
|
| 309 |
+
"epoch": 3.16,
|
| 310 |
+
"learning_rate": 3.780864197530865e-05,
|
| 311 |
+
"loss": 0.1285,
|
| 312 |
"step": 230
|
| 313 |
},
|
| 314 |
{
|
| 315 |
+
"epoch": 3.23,
|
| 316 |
+
"learning_rate": 3.742283950617284e-05,
|
| 317 |
+
"loss": 0.0848,
|
| 318 |
"step": 235
|
| 319 |
},
|
| 320 |
{
|
| 321 |
+
"epoch": 3.3,
|
| 322 |
+
"learning_rate": 3.7037037037037037e-05,
|
| 323 |
+
"loss": 0.1008,
|
| 324 |
"step": 240
|
| 325 |
},
|
| 326 |
{
|
| 327 |
+
"epoch": 3.37,
|
| 328 |
+
"learning_rate": 3.665123456790124e-05,
|
| 329 |
+
"loss": 0.0919,
|
| 330 |
"step": 245
|
| 331 |
},
|
| 332 |
{
|
| 333 |
+
"epoch": 3.44,
|
| 334 |
+
"learning_rate": 3.626543209876543e-05,
|
| 335 |
+
"loss": 0.0997,
|
| 336 |
"step": 250
|
| 337 |
},
|
| 338 |
{
|
| 339 |
+
"epoch": 3.51,
|
| 340 |
+
"learning_rate": 3.587962962962963e-05,
|
| 341 |
+
"loss": 0.0777,
|
| 342 |
"step": 255
|
| 343 |
},
|
| 344 |
{
|
| 345 |
+
"epoch": 3.57,
|
| 346 |
+
"learning_rate": 3.5493827160493834e-05,
|
| 347 |
+
"loss": 0.1069,
|
| 348 |
"step": 260
|
| 349 |
},
|
| 350 |
{
|
| 351 |
+
"epoch": 3.64,
|
| 352 |
+
"learning_rate": 3.5108024691358025e-05,
|
| 353 |
+
"loss": 0.1284,
|
| 354 |
"step": 265
|
| 355 |
},
|
| 356 |
{
|
| 357 |
+
"epoch": 3.71,
|
| 358 |
+
"learning_rate": 3.472222222222222e-05,
|
| 359 |
+
"loss": 0.1037,
|
| 360 |
"step": 270
|
| 361 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 362 |
{
|
| 363 |
"epoch": 3.78,
|
| 364 |
+
"learning_rate": 3.4336419753086427e-05,
|
| 365 |
+
"loss": 0.0799,
|
| 366 |
+
"step": 275
|
| 367 |
},
|
| 368 |
{
|
| 369 |
"epoch": 3.85,
|
| 370 |
+
"learning_rate": 3.395061728395062e-05,
|
| 371 |
+
"loss": 0.1172,
|
| 372 |
+
"step": 280
|
| 373 |
},
|
| 374 |
{
|
| 375 |
"epoch": 3.92,
|
| 376 |
+
"learning_rate": 3.3564814814814815e-05,
|
| 377 |
+
"loss": 0.1136,
|
| 378 |
+
"step": 285
|
| 379 |
},
|
| 380 |
{
|
| 381 |
"epoch": 3.99,
|
| 382 |
+
"learning_rate": 3.317901234567901e-05,
|
| 383 |
+
"loss": 0.1065,
|
| 384 |
+
"step": 290
|
| 385 |
},
|
| 386 |
{
|
| 387 |
"epoch": 4.0,
|
| 388 |
+
"eval_accuracy": 0.9777131782945736,
|
| 389 |
+
"eval_loss": 0.0747998058795929,
|
| 390 |
+
"eval_runtime": 6.5479,
|
| 391 |
+
"eval_samples_per_second": 157.608,
|
| 392 |
+
"eval_steps_per_second": 5.04,
|
| 393 |
+
"step": 291
|
| 394 |
},
|
| 395 |
{
|
| 396 |
"epoch": 4.05,
|
| 397 |
+
"learning_rate": 3.279320987654321e-05,
|
| 398 |
+
"loss": 0.0848,
|
| 399 |
+
"step": 295
|
| 400 |
},
|
| 401 |
{
|
| 402 |
"epoch": 4.12,
|
| 403 |
+
"learning_rate": 3.240740740740741e-05,
|
| 404 |
+
"loss": 0.1073,
|
| 405 |
+
"step": 300
|
| 406 |
},
|
| 407 |
{
|
| 408 |
"epoch": 4.19,
|
| 409 |
+
"learning_rate": 3.2021604938271605e-05,
|
| 410 |
+
"loss": 0.0821,
|
| 411 |
+
"step": 305
|
| 412 |
},
|
| 413 |
{
|
| 414 |
"epoch": 4.26,
|
| 415 |
+
"learning_rate": 3.16358024691358e-05,
|
| 416 |
+
"loss": 0.0662,
|
| 417 |
+
"step": 310
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 4.33,
|
| 421 |
+
"learning_rate": 3.125e-05,
|
| 422 |
+
"loss": 0.123,
|
| 423 |
"step": 315
|
| 424 |
},
|
| 425 |
{
|
| 426 |
+
"epoch": 4.4,
|
| 427 |
+
"learning_rate": 3.08641975308642e-05,
|
| 428 |
+
"loss": 0.1168,
|
| 429 |
"step": 320
|
| 430 |
},
|
| 431 |
{
|
| 432 |
+
"epoch": 4.47,
|
| 433 |
+
"learning_rate": 3.04783950617284e-05,
|
| 434 |
+
"loss": 0.0906,
|
| 435 |
"step": 325
|
| 436 |
},
|
| 437 |
{
|
| 438 |
+
"epoch": 4.54,
|
| 439 |
+
"learning_rate": 3.0092592592592593e-05,
|
| 440 |
+
"loss": 0.0801,
|
| 441 |
"step": 330
|
| 442 |
},
|
| 443 |
{
|
| 444 |
+
"epoch": 4.6,
|
| 445 |
+
"learning_rate": 2.970679012345679e-05,
|
| 446 |
+
"loss": 0.0894,
|
| 447 |
"step": 335
|
| 448 |
},
|
| 449 |
{
|
| 450 |
+
"epoch": 4.67,
|
| 451 |
+
"learning_rate": 2.9320987654320992e-05,
|
| 452 |
+
"loss": 0.0927,
|
| 453 |
"step": 340
|
| 454 |
},
|
| 455 |
{
|
| 456 |
+
"epoch": 4.74,
|
| 457 |
+
"learning_rate": 2.8935185185185186e-05,
|
| 458 |
+
"loss": 0.0803,
|
| 459 |
"step": 345
|
| 460 |
},
|
| 461 |
{
|
| 462 |
+
"epoch": 4.81,
|
| 463 |
+
"learning_rate": 2.8549382716049384e-05,
|
| 464 |
+
"loss": 0.0642,
|
| 465 |
"step": 350
|
| 466 |
},
|
| 467 |
{
|
| 468 |
+
"epoch": 4.88,
|
| 469 |
+
"learning_rate": 2.8163580246913578e-05,
|
| 470 |
+
"loss": 0.1422,
|
| 471 |
"step": 355
|
| 472 |
},
|
| 473 |
{
|
| 474 |
+
"epoch": 4.95,
|
| 475 |
+
"learning_rate": 2.777777777777778e-05,
|
| 476 |
+
"loss": 0.1115,
|
| 477 |
"step": 360
|
| 478 |
},
|
| 479 |
{
|
| 480 |
+
"epoch": 4.99,
|
| 481 |
+
"eval_accuracy": 0.9728682170542635,
|
| 482 |
+
"eval_loss": 0.09473349153995514,
|
| 483 |
+
"eval_runtime": 6.8174,
|
| 484 |
+
"eval_samples_per_second": 151.376,
|
| 485 |
+
"eval_steps_per_second": 4.841,
|
| 486 |
+
"step": 363
|
| 487 |
},
|
| 488 |
{
|
| 489 |
+
"epoch": 5.02,
|
| 490 |
+
"learning_rate": 2.7391975308641977e-05,
|
| 491 |
+
"loss": 0.0944,
|
| 492 |
+
"step": 365
|
| 493 |
},
|
| 494 |
{
|
| 495 |
+
"epoch": 5.09,
|
| 496 |
+
"learning_rate": 2.700617283950617e-05,
|
| 497 |
+
"loss": 0.0553,
|
|
|
|
|
|
|
|
|
|
| 498 |
"step": 370
|
| 499 |
},
|
| 500 |
{
|
| 501 |
+
"epoch": 5.15,
|
| 502 |
+
"learning_rate": 2.6620370370370372e-05,
|
| 503 |
+
"loss": 0.0994,
|
| 504 |
"step": 375
|
| 505 |
},
|
| 506 |
{
|
| 507 |
+
"epoch": 5.22,
|
| 508 |
+
"learning_rate": 2.623456790123457e-05,
|
| 509 |
+
"loss": 0.0802,
|
| 510 |
"step": 380
|
| 511 |
},
|
| 512 |
{
|
| 513 |
+
"epoch": 5.29,
|
| 514 |
+
"learning_rate": 2.5848765432098764e-05,
|
| 515 |
+
"loss": 0.1078,
|
| 516 |
"step": 385
|
| 517 |
},
|
| 518 |
{
|
| 519 |
+
"epoch": 5.36,
|
| 520 |
+
"learning_rate": 2.5462962962962965e-05,
|
| 521 |
+
"loss": 0.0746,
|
| 522 |
"step": 390
|
| 523 |
},
|
| 524 |
{
|
| 525 |
+
"epoch": 5.43,
|
| 526 |
+
"learning_rate": 2.5077160493827162e-05,
|
| 527 |
+
"loss": 0.0987,
|
| 528 |
"step": 395
|
| 529 |
},
|
| 530 |
{
|
| 531 |
+
"epoch": 5.5,
|
| 532 |
+
"learning_rate": 2.4691358024691357e-05,
|
| 533 |
+
"loss": 0.0885,
|
| 534 |
"step": 400
|
| 535 |
},
|
| 536 |
{
|
| 537 |
+
"epoch": 5.57,
|
| 538 |
+
"learning_rate": 2.4305555555555558e-05,
|
| 539 |
+
"loss": 0.0873,
|
| 540 |
"step": 405
|
| 541 |
},
|
| 542 |
{
|
| 543 |
+
"epoch": 5.64,
|
| 544 |
+
"learning_rate": 2.3919753086419755e-05,
|
| 545 |
+
"loss": 0.0859,
|
| 546 |
"step": 410
|
| 547 |
},
|
| 548 |
{
|
| 549 |
+
"epoch": 5.7,
|
| 550 |
+
"learning_rate": 2.353395061728395e-05,
|
| 551 |
+
"loss": 0.0871,
|
| 552 |
"step": 415
|
| 553 |
},
|
| 554 |
{
|
| 555 |
+
"epoch": 5.77,
|
| 556 |
+
"learning_rate": 2.314814814814815e-05,
|
| 557 |
+
"loss": 0.0847,
|
| 558 |
"step": 420
|
| 559 |
},
|
| 560 |
{
|
| 561 |
+
"epoch": 5.84,
|
| 562 |
+
"learning_rate": 2.2762345679012348e-05,
|
| 563 |
+
"loss": 0.1074,
|
| 564 |
"step": 425
|
| 565 |
},
|
| 566 |
{
|
| 567 |
+
"epoch": 5.91,
|
| 568 |
+
"learning_rate": 2.2376543209876542e-05,
|
| 569 |
+
"loss": 0.1017,
|
| 570 |
"step": 430
|
| 571 |
},
|
| 572 |
{
|
| 573 |
+
"epoch": 5.98,
|
| 574 |
+
"learning_rate": 2.1990740740740743e-05,
|
| 575 |
+
"loss": 0.0804,
|
| 576 |
"step": 435
|
| 577 |
},
|
| 578 |
{
|
| 579 |
+
"epoch": 5.99,
|
| 580 |
+
"eval_accuracy": 0.9757751937984496,
|
| 581 |
+
"eval_loss": 0.08881077915430069,
|
| 582 |
+
"eval_runtime": 7.2471,
|
| 583 |
+
"eval_samples_per_second": 142.401,
|
| 584 |
+
"eval_steps_per_second": 4.554,
|
| 585 |
+
"step": 436
|
| 586 |
},
|
| 587 |
{
|
| 588 |
+
"epoch": 6.05,
|
| 589 |
+
"learning_rate": 2.1604938271604937e-05,
|
| 590 |
+
"loss": 0.0592,
|
| 591 |
+
"step": 440
|
|
|
|
|
|
|
|
|
|
| 592 |
},
|
| 593 |
{
|
| 594 |
+
"epoch": 6.12,
|
| 595 |
+
"learning_rate": 2.1219135802469135e-05,
|
| 596 |
+
"loss": 0.1185,
|
| 597 |
"step": 445
|
| 598 |
},
|
| 599 |
{
|
| 600 |
+
"epoch": 6.19,
|
| 601 |
+
"learning_rate": 2.0833333333333336e-05,
|
| 602 |
+
"loss": 0.0868,
|
| 603 |
"step": 450
|
| 604 |
},
|
| 605 |
{
|
| 606 |
+
"epoch": 6.25,
|
| 607 |
+
"learning_rate": 2.044753086419753e-05,
|
| 608 |
+
"loss": 0.0883,
|
| 609 |
"step": 455
|
| 610 |
},
|
| 611 |
{
|
| 612 |
+
"epoch": 6.32,
|
| 613 |
+
"learning_rate": 2.006172839506173e-05,
|
| 614 |
+
"loss": 0.0835,
|
| 615 |
"step": 460
|
| 616 |
},
|
| 617 |
{
|
| 618 |
+
"epoch": 6.39,
|
| 619 |
+
"learning_rate": 1.967592592592593e-05,
|
| 620 |
+
"loss": 0.0781,
|
| 621 |
"step": 465
|
| 622 |
},
|
| 623 |
{
|
| 624 |
+
"epoch": 6.46,
|
| 625 |
+
"learning_rate": 1.9290123456790123e-05,
|
| 626 |
+
"loss": 0.1039,
|
| 627 |
"step": 470
|
| 628 |
},
|
| 629 |
{
|
| 630 |
+
"epoch": 6.53,
|
| 631 |
+
"learning_rate": 1.8904320987654324e-05,
|
| 632 |
+
"loss": 0.0823,
|
| 633 |
"step": 475
|
| 634 |
},
|
| 635 |
{
|
| 636 |
+
"epoch": 6.6,
|
| 637 |
+
"learning_rate": 1.8518518518518518e-05,
|
| 638 |
+
"loss": 0.0842,
|
| 639 |
"step": 480
|
| 640 |
},
|
| 641 |
{
|
| 642 |
+
"epoch": 6.67,
|
| 643 |
+
"learning_rate": 1.8132716049382716e-05,
|
| 644 |
+
"loss": 0.0859,
|
| 645 |
"step": 485
|
| 646 |
},
|
| 647 |
{
|
| 648 |
+
"epoch": 6.74,
|
| 649 |
+
"learning_rate": 1.7746913580246917e-05,
|
| 650 |
+
"loss": 0.0751,
|
| 651 |
"step": 490
|
| 652 |
},
|
| 653 |
{
|
| 654 |
+
"epoch": 6.8,
|
| 655 |
+
"learning_rate": 1.736111111111111e-05,
|
| 656 |
+
"loss": 0.081,
|
| 657 |
"step": 495
|
| 658 |
},
|
| 659 |
{
|
| 660 |
+
"epoch": 6.87,
|
| 661 |
+
"learning_rate": 1.697530864197531e-05,
|
| 662 |
+
"loss": 0.1262,
|
| 663 |
"step": 500
|
| 664 |
},
|
| 665 |
{
|
| 666 |
+
"epoch": 6.94,
|
| 667 |
+
"learning_rate": 1.6589506172839506e-05,
|
| 668 |
+
"loss": 0.0722,
|
| 669 |
"step": 505
|
| 670 |
},
|
| 671 |
{
|
| 672 |
+
"epoch": 7.0,
|
| 673 |
+
"eval_accuracy": 0.9757751937984496,
|
| 674 |
+
"eval_loss": 0.08266593515872955,
|
| 675 |
+
"eval_runtime": 7.0856,
|
| 676 |
+
"eval_samples_per_second": 145.648,
|
| 677 |
+
"eval_steps_per_second": 4.657,
|
| 678 |
+
"step": 509
|
| 679 |
},
|
| 680 |
{
|
| 681 |
+
"epoch": 7.01,
|
| 682 |
+
"learning_rate": 1.6203703703703704e-05,
|
| 683 |
+
"loss": 0.1235,
|
| 684 |
+
"step": 510
|
| 685 |
},
|
| 686 |
{
|
| 687 |
+
"epoch": 7.08,
|
| 688 |
+
"learning_rate": 1.58179012345679e-05,
|
| 689 |
+
"loss": 0.0734,
|
| 690 |
+
"step": 515
|
|
|
|
|
|
|
|
|
|
| 691 |
},
|
| 692 |
{
|
| 693 |
+
"epoch": 7.15,
|
| 694 |
+
"learning_rate": 1.54320987654321e-05,
|
| 695 |
+
"loss": 0.0819,
|
| 696 |
"step": 520
|
| 697 |
},
|
| 698 |
{
|
| 699 |
+
"epoch": 7.22,
|
| 700 |
+
"learning_rate": 1.5046296296296297e-05,
|
| 701 |
+
"loss": 0.0916,
|
| 702 |
"step": 525
|
| 703 |
},
|
| 704 |
{
|
| 705 |
+
"epoch": 7.29,
|
| 706 |
+
"learning_rate": 1.4660493827160496e-05,
|
| 707 |
+
"loss": 0.0859,
|
| 708 |
"step": 530
|
| 709 |
},
|
| 710 |
{
|
| 711 |
+
"epoch": 7.35,
|
| 712 |
+
"learning_rate": 1.4274691358024692e-05,
|
| 713 |
+
"loss": 0.1009,
|
| 714 |
"step": 535
|
| 715 |
},
|
| 716 |
{
|
| 717 |
+
"epoch": 7.42,
|
| 718 |
+
"learning_rate": 1.388888888888889e-05,
|
| 719 |
+
"loss": 0.0996,
|
| 720 |
"step": 540
|
| 721 |
},
|
| 722 |
{
|
| 723 |
+
"epoch": 7.49,
|
| 724 |
+
"learning_rate": 1.3503086419753085e-05,
|
| 725 |
+
"loss": 0.1062,
|
| 726 |
"step": 545
|
| 727 |
},
|
| 728 |
{
|
| 729 |
+
"epoch": 7.56,
|
| 730 |
+
"learning_rate": 1.3117283950617285e-05,
|
| 731 |
+
"loss": 0.0759,
|
| 732 |
"step": 550
|
| 733 |
},
|
| 734 |
{
|
| 735 |
+
"epoch": 7.63,
|
| 736 |
+
"learning_rate": 1.2731481481481482e-05,
|
| 737 |
+
"loss": 0.0658,
|
| 738 |
"step": 555
|
| 739 |
},
|
| 740 |
{
|
| 741 |
+
"epoch": 7.7,
|
| 742 |
+
"learning_rate": 1.2345679012345678e-05,
|
| 743 |
+
"loss": 0.1081,
|
| 744 |
"step": 560
|
| 745 |
},
|
| 746 |
{
|
| 747 |
+
"epoch": 7.77,
|
| 748 |
+
"learning_rate": 1.1959876543209878e-05,
|
| 749 |
+
"loss": 0.0866,
|
| 750 |
"step": 565
|
| 751 |
},
|
| 752 |
{
|
| 753 |
+
"epoch": 7.84,
|
| 754 |
+
"learning_rate": 1.1574074074074075e-05,
|
| 755 |
+
"loss": 0.0783,
|
| 756 |
"step": 570
|
| 757 |
},
|
| 758 |
{
|
| 759 |
+
"epoch": 7.9,
|
| 760 |
+
"learning_rate": 1.1188271604938271e-05,
|
| 761 |
+
"loss": 0.0868,
|
| 762 |
"step": 575
|
| 763 |
},
|
| 764 |
{
|
| 765 |
+
"epoch": 7.97,
|
| 766 |
+
"learning_rate": 1.0802469135802469e-05,
|
| 767 |
+
"loss": 0.061,
|
| 768 |
"step": 580
|
| 769 |
},
|
| 770 |
{
|
| 771 |
+
"epoch": 8.0,
|
| 772 |
+
"eval_accuracy": 0.9757751937984496,
|
| 773 |
+
"eval_loss": 0.0899147242307663,
|
| 774 |
+
"eval_runtime": 6.5821,
|
| 775 |
+
"eval_samples_per_second": 156.788,
|
| 776 |
+
"eval_steps_per_second": 5.014,
|
| 777 |
+
"step": 582
|
| 778 |
},
|
| 779 |
{
|
| 780 |
+
"epoch": 8.04,
|
| 781 |
+
"learning_rate": 1.0416666666666668e-05,
|
| 782 |
+
"loss": 0.053,
|
| 783 |
+
"step": 585
|
| 784 |
},
|
| 785 |
{
|
| 786 |
+
"epoch": 8.11,
|
| 787 |
+
"learning_rate": 1.0030864197530866e-05,
|
| 788 |
+
"loss": 0.0749,
|
| 789 |
+
"step": 590
|
|
|
|
|
|
|
|
|
|
| 790 |
},
|
| 791 |
{
|
| 792 |
+
"epoch": 8.18,
|
| 793 |
+
"learning_rate": 9.645061728395062e-06,
|
| 794 |
+
"loss": 0.0607,
|
| 795 |
"step": 595
|
| 796 |
},
|
| 797 |
{
|
| 798 |
+
"epoch": 8.25,
|
| 799 |
+
"learning_rate": 9.259259259259259e-06,
|
| 800 |
+
"loss": 0.0772,
|
| 801 |
"step": 600
|
| 802 |
},
|
| 803 |
{
|
| 804 |
+
"epoch": 8.32,
|
| 805 |
+
"learning_rate": 8.873456790123458e-06,
|
| 806 |
+
"loss": 0.055,
|
| 807 |
"step": 605
|
| 808 |
},
|
| 809 |
{
|
| 810 |
+
"epoch": 8.38,
|
| 811 |
+
"learning_rate": 8.487654320987654e-06,
|
| 812 |
+
"loss": 0.0647,
|
| 813 |
"step": 610
|
| 814 |
},
|
| 815 |
{
|
| 816 |
+
"epoch": 8.45,
|
| 817 |
+
"learning_rate": 8.101851851851852e-06,
|
| 818 |
+
"loss": 0.1082,
|
| 819 |
"step": 615
|
| 820 |
},
|
| 821 |
{
|
| 822 |
+
"epoch": 8.52,
|
| 823 |
+
"learning_rate": 7.71604938271605e-06,
|
| 824 |
+
"loss": 0.09,
|
| 825 |
"step": 620
|
| 826 |
},
|
| 827 |
{
|
| 828 |
+
"epoch": 8.59,
|
| 829 |
+
"learning_rate": 7.330246913580248e-06,
|
| 830 |
+
"loss": 0.0739,
|
| 831 |
"step": 625
|
| 832 |
},
|
| 833 |
{
|
| 834 |
+
"epoch": 8.66,
|
| 835 |
+
"learning_rate": 6.944444444444445e-06,
|
| 836 |
+
"loss": 0.0632,
|
| 837 |
"step": 630
|
| 838 |
},
|
| 839 |
{
|
| 840 |
+
"epoch": 8.73,
|
| 841 |
+
"learning_rate": 6.558641975308642e-06,
|
| 842 |
+
"loss": 0.0662,
|
| 843 |
"step": 635
|
| 844 |
},
|
| 845 |
{
|
| 846 |
+
"epoch": 8.8,
|
| 847 |
+
"learning_rate": 6.172839506172839e-06,
|
| 848 |
+
"loss": 0.0691,
|
| 849 |
"step": 640
|
| 850 |
},
|
| 851 |
{
|
| 852 |
+
"epoch": 8.87,
|
| 853 |
+
"learning_rate": 5.787037037037038e-06,
|
| 854 |
+
"loss": 0.0745,
|
| 855 |
"step": 645
|
| 856 |
},
|
| 857 |
{
|
| 858 |
+
"epoch": 8.93,
|
| 859 |
+
"learning_rate": 5.401234567901234e-06,
|
| 860 |
+
"loss": 0.0706,
|
| 861 |
"step": 650
|
| 862 |
},
|
| 863 |
{
|
| 864 |
+
"epoch": 8.99,
|
| 865 |
+
"eval_accuracy": 0.9757751937984496,
|
| 866 |
+
"eval_loss": 0.09157940745353699,
|
| 867 |
+
"eval_runtime": 6.6599,
|
| 868 |
+
"eval_samples_per_second": 154.957,
|
| 869 |
+
"eval_steps_per_second": 4.955,
|
| 870 |
+
"step": 654
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 9.0,
|
| 874 |
+
"learning_rate": 5.015432098765433e-06,
|
| 875 |
+
"loss": 0.0871,
|
| 876 |
"step": 655
|
| 877 |
},
|
| 878 |
{
|
| 879 |
+
"epoch": 9.07,
|
| 880 |
+
"learning_rate": 4.6296296296296296e-06,
|
| 881 |
+
"loss": 0.101,
|
| 882 |
"step": 660
|
| 883 |
},
|
| 884 |
{
|
| 885 |
+
"epoch": 9.14,
|
| 886 |
+
"learning_rate": 4.243827160493827e-06,
|
| 887 |
+
"loss": 0.0738,
|
| 888 |
"step": 665
|
| 889 |
},
|
| 890 |
{
|
| 891 |
+
"epoch": 9.21,
|
| 892 |
+
"learning_rate": 3.858024691358025e-06,
|
| 893 |
+
"loss": 0.0803,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 894 |
"step": 670
|
| 895 |
},
|
| 896 |
{
|
| 897 |
+
"epoch": 9.28,
|
| 898 |
+
"learning_rate": 3.4722222222222224e-06,
|
| 899 |
+
"loss": 0.0787,
|
| 900 |
"step": 675
|
| 901 |
},
|
| 902 |
{
|
| 903 |
+
"epoch": 9.35,
|
| 904 |
+
"learning_rate": 3.0864197530864196e-06,
|
| 905 |
+
"loss": 0.0661,
|
| 906 |
"step": 680
|
| 907 |
},
|
| 908 |
{
|
| 909 |
+
"epoch": 9.42,
|
| 910 |
+
"learning_rate": 2.700617283950617e-06,
|
| 911 |
+
"loss": 0.055,
|
| 912 |
"step": 685
|
| 913 |
},
|
| 914 |
{
|
| 915 |
+
"epoch": 9.48,
|
| 916 |
+
"learning_rate": 2.3148148148148148e-06,
|
| 917 |
+
"loss": 0.0752,
|
| 918 |
"step": 690
|
| 919 |
},
|
| 920 |
{
|
| 921 |
+
"epoch": 9.55,
|
| 922 |
+
"learning_rate": 1.9290123456790124e-06,
|
| 923 |
+
"loss": 0.0933,
|
| 924 |
"step": 695
|
| 925 |
},
|
| 926 |
{
|
| 927 |
+
"epoch": 9.62,
|
| 928 |
+
"learning_rate": 1.5432098765432098e-06,
|
| 929 |
+
"loss": 0.1046,
|
| 930 |
"step": 700
|
| 931 |
},
|
| 932 |
{
|
| 933 |
+
"epoch": 9.69,
|
| 934 |
+
"learning_rate": 1.1574074074074074e-06,
|
| 935 |
+
"loss": 0.0467,
|
| 936 |
"step": 705
|
| 937 |
},
|
| 938 |
{
|
| 939 |
+
"epoch": 9.76,
|
| 940 |
+
"learning_rate": 7.716049382716049e-07,
|
| 941 |
+
"loss": 0.0739,
|
| 942 |
"step": 710
|
| 943 |
},
|
| 944 |
{
|
| 945 |
+
"epoch": 9.83,
|
| 946 |
+
"learning_rate": 3.8580246913580245e-07,
|
| 947 |
+
"loss": 0.068,
|
| 948 |
"step": 715
|
| 949 |
},
|
| 950 |
{
|
| 951 |
+
"epoch": 9.9,
|
| 952 |
+
"learning_rate": 0.0,
|
| 953 |
+
"loss": 0.0633,
|
| 954 |
"step": 720
|
| 955 |
},
|
| 956 |
{
|
| 957 |
+
"epoch": 9.9,
|
| 958 |
+
"eval_accuracy": 0.9757751937984496,
|
| 959 |
+
"eval_loss": 0.09366881847381592,
|
| 960 |
+
"eval_runtime": 6.8299,
|
| 961 |
+
"eval_samples_per_second": 151.1,
|
| 962 |
+
"eval_steps_per_second": 4.832,
|
| 963 |
+
"step": 720
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 964 |
},
|
| 965 |
{
|
| 966 |
+
"epoch": 9.9,
|
| 967 |
+
"step": 720,
|
| 968 |
+
"total_flos": 2.2837931240094351e+18,
|
| 969 |
+
"train_loss": 0.11106966810507907,
|
| 970 |
+
"train_runtime": 1435.0993,
|
| 971 |
+
"train_samples_per_second": 64.671,
|
| 972 |
+
"train_steps_per_second": 0.502
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 973 |
}
|
| 974 |
],
|
| 975 |
"logging_steps": 5,
|
| 976 |
+
"max_steps": 720,
|
| 977 |
"num_train_epochs": 10,
|
| 978 |
"save_steps": 500,
|
| 979 |
+
"total_flos": 2.2837931240094351e+18,
|
| 980 |
"trial_name": null,
|
| 981 |
"trial_params": null
|
| 982 |
}
|