Model save
Browse files- README.md +7 -7
- all_results.json +5 -5
- test_results.json +5 -5
- trainer_state.json +229 -229
- val_results.json +5 -5
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 [facebook/timesformer-base-finetuned-k400](https://huggingface.co/facebook/timesformer-base-finetuned-k400) 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 |
|
|
@@ -50,11 +50,11 @@ The following hyperparameters were used during training:
|
|
| 50 |
|
| 51 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 52 |
|:-------------:|:------:|:----:|:---------------:|:--------:|
|
| 53 |
-
| 0.
|
| 54 |
-
| 0.
|
| 55 |
-
| 0.
|
| 56 |
-
| 0.
|
| 57 |
-
| 0.
|
| 58 |
|
| 59 |
|
| 60 |
### Framework versions
|
|
|
|
| 17 |
|
| 18 |
This model is a fine-tuned version of [facebook/timesformer-base-finetuned-k400](https://huggingface.co/facebook/timesformer-base-finetuned-k400) on an unknown dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 0.6192
|
| 21 |
+
- Accuracy: 0.8382
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
|
|
| 50 |
|
| 51 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 52 |
|:-------------:|:------:|:----:|:---------------:|:--------:|
|
| 53 |
+
| 0.4491 | 0.2011 | 186 | 0.6939 | 0.7386 |
|
| 54 |
+
| 0.5627 | 1.2011 | 372 | 0.6806 | 0.7759 |
|
| 55 |
+
| 0.5189 | 2.2011 | 558 | 0.6510 | 0.8174 |
|
| 56 |
+
| 0.2503 | 3.2011 | 744 | 0.6732 | 0.8174 |
|
| 57 |
+
| 0.0159 | 4.1957 | 925 | 0.6192 | 0.8382 |
|
| 58 |
|
| 59 |
|
| 60 |
### Framework versions
|
all_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 4.195675675675676,
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second": 0.
|
| 7 |
-
"eval_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 4.195675675675676,
|
| 3 |
+
"eval_accuracy": 0.8381742738589212,
|
| 4 |
+
"eval_loss": 0.6192311644554138,
|
| 5 |
+
"eval_runtime": 391.274,
|
| 6 |
+
"eval_samples_per_second": 0.616,
|
| 7 |
+
"eval_steps_per_second": 0.156
|
| 8 |
}
|
test_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 4.195675675675676,
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second": 0.
|
| 7 |
-
"eval_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 4.195675675675676,
|
| 3 |
+
"eval_accuracy": 0.8134831460674158,
|
| 4 |
+
"eval_loss": 0.7141955494880676,
|
| 5 |
+
"eval_runtime": 720.0647,
|
| 6 |
+
"eval_samples_per_second": 0.618,
|
| 7 |
+
"eval_steps_per_second": 0.156
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,5 +1,5 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
"best_model_checkpoint": "videomae-timesformer-surf-analytics\\checkpoint-925",
|
| 4 |
"epoch": 4.195675675675676,
|
| 5 |
"eval_steps": 500,
|
|
@@ -10,727 +10,727 @@
|
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.010810810810810811,
|
| 13 |
-
"grad_norm":
|
| 14 |
"learning_rate": 5.376344086021506e-06,
|
| 15 |
-
"loss": 1.
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.021621621621621623,
|
| 20 |
-
"grad_norm": 11.
|
| 21 |
"learning_rate": 1.0752688172043012e-05,
|
| 22 |
-
"loss": 1.
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.032432432432432434,
|
| 27 |
-
"grad_norm":
|
| 28 |
"learning_rate": 1.6129032258064517e-05,
|
| 29 |
-
"loss": 1.
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.043243243243243246,
|
| 34 |
-
"grad_norm":
|
| 35 |
"learning_rate": 2.1505376344086024e-05,
|
| 36 |
-
"loss": 1.
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.05405405405405406,
|
| 41 |
-
"grad_norm":
|
| 42 |
"learning_rate": 2.6881720430107527e-05,
|
| 43 |
-
"loss": 1.
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.06486486486486487,
|
| 48 |
-
"grad_norm":
|
| 49 |
"learning_rate": 3.2258064516129034e-05,
|
| 50 |
-
"loss":
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.07567567567567568,
|
| 55 |
-
"grad_norm":
|
| 56 |
"learning_rate": 3.763440860215054e-05,
|
| 57 |
-
"loss": 1.
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.08648648648648649,
|
| 62 |
-
"grad_norm":
|
| 63 |
"learning_rate": 4.301075268817205e-05,
|
| 64 |
-
"loss": 0.
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.0972972972972973,
|
| 69 |
-
"grad_norm":
|
| 70 |
"learning_rate": 4.8387096774193554e-05,
|
| 71 |
-
"loss": 0.
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 0.10810810810810811,
|
| 76 |
-
"grad_norm":
|
| 77 |
"learning_rate": 4.957932692307692e-05,
|
| 78 |
-
"loss": 0.
|
| 79 |
"step": 100
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 0.11891891891891893,
|
| 83 |
-
"grad_norm": 7.
|
| 84 |
"learning_rate": 4.897836538461539e-05,
|
| 85 |
-
"loss": 0.
|
| 86 |
"step": 110
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 0.12972972972972974,
|
| 90 |
-
"grad_norm":
|
| 91 |
"learning_rate": 4.8377403846153846e-05,
|
| 92 |
-
"loss": 0.
|
| 93 |
"step": 120
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 0.14054054054054055,
|
| 97 |
-
"grad_norm":
|
| 98 |
"learning_rate": 4.777644230769231e-05,
|
| 99 |
-
"loss": 0.
|
| 100 |
"step": 130
|
| 101 |
},
|
| 102 |
{
|
| 103 |
"epoch": 0.15135135135135136,
|
| 104 |
-
"grad_norm":
|
| 105 |
"learning_rate": 4.717548076923077e-05,
|
| 106 |
-
"loss": 0.
|
| 107 |
"step": 140
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 0.16216216216216217,
|
| 111 |
-
"grad_norm":
|
| 112 |
"learning_rate": 4.657451923076923e-05,
|
| 113 |
-
"loss": 0.
|
| 114 |
"step": 150
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 0.17297297297297298,
|
| 118 |
-
"grad_norm":
|
| 119 |
"learning_rate": 4.5973557692307694e-05,
|
| 120 |
-
"loss": 1.
|
| 121 |
"step": 160
|
| 122 |
},
|
| 123 |
{
|
| 124 |
"epoch": 0.1837837837837838,
|
| 125 |
-
"grad_norm":
|
| 126 |
"learning_rate": 4.5372596153846156e-05,
|
| 127 |
-
"loss": 0.
|
| 128 |
"step": 170
|
| 129 |
},
|
| 130 |
{
|
| 131 |
"epoch": 0.1945945945945946,
|
| 132 |
-
"grad_norm":
|
| 133 |
"learning_rate": 4.477163461538462e-05,
|
| 134 |
-
"loss": 0.
|
| 135 |
"step": 180
|
| 136 |
},
|
| 137 |
{
|
| 138 |
"epoch": 0.20108108108108108,
|
| 139 |
-
"eval_accuracy": 0.
|
| 140 |
-
"eval_loss": 0.
|
| 141 |
-
"eval_runtime":
|
| 142 |
-
"eval_samples_per_second": 0.
|
| 143 |
-
"eval_steps_per_second": 0.
|
| 144 |
"step": 186
|
| 145 |
},
|
| 146 |
{
|
| 147 |
"epoch": 1.0043243243243243,
|
| 148 |
-
"grad_norm":
|
| 149 |
"learning_rate": 4.417067307692308e-05,
|
| 150 |
-
"loss": 0.
|
| 151 |
"step": 190
|
| 152 |
},
|
| 153 |
{
|
| 154 |
"epoch": 1.0151351351351352,
|
| 155 |
-
"grad_norm":
|
| 156 |
"learning_rate": 4.3569711538461535e-05,
|
| 157 |
-
"loss": 0.
|
| 158 |
"step": 200
|
| 159 |
},
|
| 160 |
{
|
| 161 |
"epoch": 1.025945945945946,
|
| 162 |
-
"grad_norm":
|
| 163 |
"learning_rate": 4.2968750000000004e-05,
|
| 164 |
-
"loss": 0.
|
| 165 |
"step": 210
|
| 166 |
},
|
| 167 |
{
|
| 168 |
"epoch": 1.0367567567567568,
|
| 169 |
-
"grad_norm":
|
| 170 |
"learning_rate": 4.2367788461538466e-05,
|
| 171 |
-
"loss": 0.
|
| 172 |
"step": 220
|
| 173 |
},
|
| 174 |
{
|
| 175 |
"epoch": 1.0475675675675675,
|
| 176 |
-
"grad_norm":
|
| 177 |
"learning_rate": 4.176682692307692e-05,
|
| 178 |
-
"loss": 0.
|
| 179 |
"step": 230
|
| 180 |
},
|
| 181 |
{
|
| 182 |
"epoch": 1.0583783783783784,
|
| 183 |
-
"grad_norm":
|
| 184 |
"learning_rate": 4.116586538461539e-05,
|
| 185 |
-
"loss": 0.
|
| 186 |
"step": 240
|
| 187 |
},
|
| 188 |
{
|
| 189 |
"epoch": 1.0691891891891891,
|
| 190 |
-
"grad_norm":
|
| 191 |
"learning_rate": 4.0564903846153846e-05,
|
| 192 |
-
"loss": 0.
|
| 193 |
"step": 250
|
| 194 |
},
|
| 195 |
{
|
| 196 |
"epoch": 1.08,
|
| 197 |
-
"grad_norm":
|
| 198 |
"learning_rate": 3.996394230769231e-05,
|
| 199 |
-
"loss": 0.
|
| 200 |
"step": 260
|
| 201 |
},
|
| 202 |
{
|
| 203 |
"epoch": 1.0908108108108108,
|
| 204 |
-
"grad_norm":
|
| 205 |
"learning_rate": 3.936298076923077e-05,
|
| 206 |
-
"loss": 0.
|
| 207 |
"step": 270
|
| 208 |
},
|
| 209 |
{
|
| 210 |
"epoch": 1.1016216216216217,
|
| 211 |
-
"grad_norm":
|
| 212 |
"learning_rate": 3.876201923076923e-05,
|
| 213 |
-
"loss": 0.
|
| 214 |
"step": 280
|
| 215 |
},
|
| 216 |
{
|
| 217 |
"epoch": 1.1124324324324324,
|
| 218 |
-
"grad_norm":
|
| 219 |
"learning_rate": 3.8161057692307694e-05,
|
| 220 |
-
"loss": 0.
|
| 221 |
"step": 290
|
| 222 |
},
|
| 223 |
{
|
| 224 |
"epoch": 1.1232432432432433,
|
| 225 |
-
"grad_norm":
|
| 226 |
"learning_rate": 3.7560096153846156e-05,
|
| 227 |
-
"loss": 0.
|
| 228 |
"step": 300
|
| 229 |
},
|
| 230 |
{
|
| 231 |
"epoch": 1.134054054054054,
|
| 232 |
-
"grad_norm":
|
| 233 |
"learning_rate": 3.695913461538462e-05,
|
| 234 |
-
"loss": 0.
|
| 235 |
"step": 310
|
| 236 |
},
|
| 237 |
{
|
| 238 |
"epoch": 1.144864864864865,
|
| 239 |
-
"grad_norm":
|
| 240 |
"learning_rate": 3.635817307692308e-05,
|
| 241 |
-
"loss": 0.
|
| 242 |
"step": 320
|
| 243 |
},
|
| 244 |
{
|
| 245 |
"epoch": 1.1556756756756756,
|
| 246 |
-
"grad_norm":
|
| 247 |
"learning_rate": 3.5757211538461535e-05,
|
| 248 |
-
"loss": 0.
|
| 249 |
"step": 330
|
| 250 |
},
|
| 251 |
{
|
| 252 |
"epoch": 1.1664864864864866,
|
| 253 |
-
"grad_norm": 0.
|
| 254 |
"learning_rate": 3.5156250000000004e-05,
|
| 255 |
-
"loss": 0.
|
| 256 |
"step": 340
|
| 257 |
},
|
| 258 |
{
|
| 259 |
"epoch": 1.1772972972972973,
|
| 260 |
-
"grad_norm": 0.
|
| 261 |
"learning_rate": 3.4555288461538466e-05,
|
| 262 |
-
"loss": 0.
|
| 263 |
"step": 350
|
| 264 |
},
|
| 265 |
{
|
| 266 |
"epoch": 1.1881081081081082,
|
| 267 |
-
"grad_norm":
|
| 268 |
"learning_rate": 3.395432692307692e-05,
|
| 269 |
-
"loss": 0.
|
| 270 |
"step": 360
|
| 271 |
},
|
| 272 |
{
|
| 273 |
"epoch": 1.1989189189189189,
|
| 274 |
-
"grad_norm":
|
| 275 |
"learning_rate": 3.335336538461539e-05,
|
| 276 |
-
"loss": 0.
|
| 277 |
"step": 370
|
| 278 |
},
|
| 279 |
{
|
| 280 |
"epoch": 1.201081081081081,
|
| 281 |
-
"eval_accuracy": 0.
|
| 282 |
-
"eval_loss": 0.
|
| 283 |
-
"eval_runtime":
|
| 284 |
-
"eval_samples_per_second": 0.
|
| 285 |
-
"eval_steps_per_second": 0.
|
| 286 |
"step": 372
|
| 287 |
},
|
| 288 |
{
|
| 289 |
"epoch": 2.0086486486486486,
|
| 290 |
-
"grad_norm": 0.
|
| 291 |
"learning_rate": 3.2752403846153846e-05,
|
| 292 |
-
"loss": 0.
|
| 293 |
"step": 380
|
| 294 |
},
|
| 295 |
{
|
| 296 |
"epoch": 2.0194594594594593,
|
| 297 |
-
"grad_norm":
|
| 298 |
"learning_rate": 3.215144230769231e-05,
|
| 299 |
-
"loss": 0.
|
| 300 |
"step": 390
|
| 301 |
},
|
| 302 |
{
|
| 303 |
"epoch": 2.0302702702702704,
|
| 304 |
-
"grad_norm":
|
| 305 |
"learning_rate": 3.155048076923077e-05,
|
| 306 |
-
"loss": 0.
|
| 307 |
"step": 400
|
| 308 |
},
|
| 309 |
{
|
| 310 |
"epoch": 2.041081081081081,
|
| 311 |
-
"grad_norm":
|
| 312 |
"learning_rate": 3.094951923076923e-05,
|
| 313 |
-
"loss": 0.
|
| 314 |
"step": 410
|
| 315 |
},
|
| 316 |
{
|
| 317 |
"epoch": 2.051891891891892,
|
| 318 |
-
"grad_norm": 0.
|
| 319 |
"learning_rate": 3.0348557692307694e-05,
|
| 320 |
-
"loss": 0.
|
| 321 |
"step": 420
|
| 322 |
},
|
| 323 |
{
|
| 324 |
"epoch": 2.0627027027027025,
|
| 325 |
-
"grad_norm":
|
| 326 |
"learning_rate": 2.974759615384616e-05,
|
| 327 |
-
"loss": 0.
|
| 328 |
"step": 430
|
| 329 |
},
|
| 330 |
{
|
| 331 |
"epoch": 2.0735135135135137,
|
| 332 |
-
"grad_norm": 0.
|
| 333 |
"learning_rate": 2.9146634615384614e-05,
|
| 334 |
-
"loss": 0.
|
| 335 |
"step": 440
|
| 336 |
},
|
| 337 |
{
|
| 338 |
"epoch": 2.0843243243243244,
|
| 339 |
-
"grad_norm":
|
| 340 |
"learning_rate": 2.854567307692308e-05,
|
| 341 |
-
"loss": 0.
|
| 342 |
"step": 450
|
| 343 |
},
|
| 344 |
{
|
| 345 |
"epoch": 2.095135135135135,
|
| 346 |
-
"grad_norm": 0.
|
| 347 |
"learning_rate": 2.794471153846154e-05,
|
| 348 |
-
"loss": 0.
|
| 349 |
"step": 460
|
| 350 |
},
|
| 351 |
{
|
| 352 |
"epoch": 2.1059459459459458,
|
| 353 |
-
"grad_norm": 0.
|
| 354 |
"learning_rate": 2.734375e-05,
|
| 355 |
-
"loss": 0.
|
| 356 |
"step": 470
|
| 357 |
},
|
| 358 |
{
|
| 359 |
"epoch": 2.116756756756757,
|
| 360 |
-
"grad_norm": 0.
|
| 361 |
"learning_rate": 2.6742788461538466e-05,
|
| 362 |
-
"loss": 0.
|
| 363 |
"step": 480
|
| 364 |
},
|
| 365 |
{
|
| 366 |
"epoch": 2.1275675675675676,
|
| 367 |
-
"grad_norm":
|
| 368 |
"learning_rate": 2.6141826923076925e-05,
|
| 369 |
-
"loss": 0.
|
| 370 |
"step": 490
|
| 371 |
},
|
| 372 |
{
|
| 373 |
"epoch": 2.1383783783783783,
|
| 374 |
-
"grad_norm": 0.
|
| 375 |
"learning_rate": 2.5540865384615387e-05,
|
| 376 |
-
"loss": 0.
|
| 377 |
"step": 500
|
| 378 |
},
|
| 379 |
{
|
| 380 |
"epoch": 2.149189189189189,
|
| 381 |
-
"grad_norm": 16.
|
| 382 |
"learning_rate": 2.493990384615385e-05,
|
| 383 |
-
"loss": 0.
|
| 384 |
"step": 510
|
| 385 |
},
|
| 386 |
{
|
| 387 |
"epoch": 2.16,
|
| 388 |
-
"grad_norm": 1.
|
| 389 |
"learning_rate": 2.4338942307692307e-05,
|
| 390 |
-
"loss": 0.
|
| 391 |
"step": 520
|
| 392 |
},
|
| 393 |
{
|
| 394 |
"epoch": 2.170810810810811,
|
| 395 |
-
"grad_norm":
|
| 396 |
"learning_rate": 2.373798076923077e-05,
|
| 397 |
-
"loss": 0.
|
| 398 |
"step": 530
|
| 399 |
},
|
| 400 |
{
|
| 401 |
"epoch": 2.1816216216216215,
|
| 402 |
-
"grad_norm": 0.
|
| 403 |
"learning_rate": 2.313701923076923e-05,
|
| 404 |
-
"loss": 0.
|
| 405 |
"step": 540
|
| 406 |
},
|
| 407 |
{
|
| 408 |
"epoch": 2.1924324324324322,
|
| 409 |
-
"grad_norm":
|
| 410 |
"learning_rate": 2.2536057692307694e-05,
|
| 411 |
-
"loss": 0.
|
| 412 |
"step": 550
|
| 413 |
},
|
| 414 |
{
|
| 415 |
"epoch": 2.2010810810810812,
|
| 416 |
-
"eval_accuracy": 0.
|
| 417 |
-
"eval_loss": 0.
|
| 418 |
-
"eval_runtime":
|
| 419 |
-
"eval_samples_per_second": 0.
|
| 420 |
-
"eval_steps_per_second": 0.
|
| 421 |
"step": 558
|
| 422 |
},
|
| 423 |
{
|
| 424 |
"epoch": 3.002162162162162,
|
| 425 |
-
"grad_norm":
|
| 426 |
"learning_rate": 2.1935096153846156e-05,
|
| 427 |
-
"loss": 0.
|
| 428 |
"step": 560
|
| 429 |
},
|
| 430 |
{
|
| 431 |
"epoch": 3.012972972972973,
|
| 432 |
-
"grad_norm":
|
| 433 |
"learning_rate": 2.1334134615384614e-05,
|
| 434 |
-
"loss": 0.
|
| 435 |
"step": 570
|
| 436 |
},
|
| 437 |
{
|
| 438 |
"epoch": 3.023783783783784,
|
| 439 |
-
"grad_norm": 0.
|
| 440 |
"learning_rate": 2.073317307692308e-05,
|
| 441 |
-
"loss": 0.
|
| 442 |
"step": 580
|
| 443 |
},
|
| 444 |
{
|
| 445 |
"epoch": 3.0345945945945947,
|
| 446 |
-
"grad_norm":
|
| 447 |
"learning_rate": 2.0132211538461542e-05,
|
| 448 |
-
"loss": 0.
|
| 449 |
"step": 590
|
| 450 |
},
|
| 451 |
{
|
| 452 |
"epoch": 3.0454054054054054,
|
| 453 |
-
"grad_norm": 0.
|
| 454 |
"learning_rate": 1.953125e-05,
|
| 455 |
-
"loss": 0.
|
| 456 |
"step": 600
|
| 457 |
},
|
| 458 |
{
|
| 459 |
"epoch": 3.056216216216216,
|
| 460 |
-
"grad_norm":
|
| 461 |
"learning_rate": 1.8930288461538462e-05,
|
| 462 |
-
"loss": 0.
|
| 463 |
"step": 610
|
| 464 |
},
|
| 465 |
{
|
| 466 |
"epoch": 3.0670270270270272,
|
| 467 |
-
"grad_norm": 0.
|
| 468 |
"learning_rate": 1.832932692307692e-05,
|
| 469 |
-
"loss": 0.
|
| 470 |
"step": 620
|
| 471 |
},
|
| 472 |
{
|
| 473 |
"epoch": 3.077837837837838,
|
| 474 |
-
"grad_norm":
|
| 475 |
"learning_rate": 1.7728365384615387e-05,
|
| 476 |
-
"loss": 0.
|
| 477 |
"step": 630
|
| 478 |
},
|
| 479 |
{
|
| 480 |
"epoch": 3.0886486486486486,
|
| 481 |
-
"grad_norm": 0.
|
| 482 |
"learning_rate": 1.712740384615385e-05,
|
| 483 |
-
"loss": 0.
|
| 484 |
"step": 640
|
| 485 |
},
|
| 486 |
{
|
| 487 |
"epoch": 3.0994594594594593,
|
| 488 |
-
"grad_norm":
|
| 489 |
"learning_rate": 1.6526442307692307e-05,
|
| 490 |
-
"loss": 0.
|
| 491 |
"step": 650
|
| 492 |
},
|
| 493 |
{
|
| 494 |
"epoch": 3.1102702702702705,
|
| 495 |
-
"grad_norm":
|
| 496 |
"learning_rate": 1.592548076923077e-05,
|
| 497 |
-
"loss": 0.
|
| 498 |
"step": 660
|
| 499 |
},
|
| 500 |
{
|
| 501 |
"epoch": 3.121081081081081,
|
| 502 |
-
"grad_norm": 0.
|
| 503 |
"learning_rate": 1.532451923076923e-05,
|
| 504 |
-
"loss": 0.
|
| 505 |
"step": 670
|
| 506 |
},
|
| 507 |
{
|
| 508 |
"epoch": 3.131891891891892,
|
| 509 |
-
"grad_norm": 0.
|
| 510 |
"learning_rate": 1.4723557692307693e-05,
|
| 511 |
-
"loss": 0.
|
| 512 |
"step": 680
|
| 513 |
},
|
| 514 |
{
|
| 515 |
"epoch": 3.1427027027027026,
|
| 516 |
-
"grad_norm":
|
| 517 |
"learning_rate": 1.4122596153846154e-05,
|
| 518 |
-
"loss": 0.
|
| 519 |
"step": 690
|
| 520 |
},
|
| 521 |
{
|
| 522 |
"epoch": 3.1535135135135137,
|
| 523 |
-
"grad_norm": 0.
|
| 524 |
"learning_rate": 1.3521634615384616e-05,
|
| 525 |
-
"loss": 0.
|
| 526 |
"step": 700
|
| 527 |
},
|
| 528 |
{
|
| 529 |
"epoch": 3.1643243243243244,
|
| 530 |
-
"grad_norm": 0.
|
| 531 |
"learning_rate": 1.292067307692308e-05,
|
| 532 |
-
"loss": 0.
|
| 533 |
"step": 710
|
| 534 |
},
|
| 535 |
{
|
| 536 |
"epoch": 3.175135135135135,
|
| 537 |
-
"grad_norm": 0.
|
| 538 |
"learning_rate": 1.231971153846154e-05,
|
| 539 |
-
"loss": 0.
|
| 540 |
"step": 720
|
| 541 |
},
|
| 542 |
{
|
| 543 |
"epoch": 3.185945945945946,
|
| 544 |
-
"grad_norm":
|
| 545 |
"learning_rate": 1.171875e-05,
|
| 546 |
-
"loss": 0.
|
| 547 |
"step": 730
|
| 548 |
},
|
| 549 |
{
|
| 550 |
"epoch": 3.1967567567567565,
|
| 551 |
-
"grad_norm": 0.
|
| 552 |
"learning_rate": 1.111778846153846e-05,
|
| 553 |
-
"loss": 0.
|
| 554 |
"step": 740
|
| 555 |
},
|
| 556 |
{
|
| 557 |
"epoch": 3.2010810810810812,
|
| 558 |
-
"eval_accuracy": 0.
|
| 559 |
-
"eval_loss": 0.
|
| 560 |
-
"eval_runtime":
|
| 561 |
-
"eval_samples_per_second": 0.
|
| 562 |
-
"eval_steps_per_second": 0.
|
| 563 |
"step": 744
|
| 564 |
},
|
| 565 |
{
|
| 566 |
"epoch": 4.006486486486486,
|
| 567 |
-
"grad_norm": 0.
|
| 568 |
"learning_rate": 1.0516826923076924e-05,
|
| 569 |
-
"loss": 0.
|
| 570 |
"step": 750
|
| 571 |
},
|
| 572 |
{
|
| 573 |
"epoch": 4.017297297297297,
|
| 574 |
-
"grad_norm":
|
| 575 |
"learning_rate": 9.915865384615385e-06,
|
| 576 |
-
"loss": 0.
|
| 577 |
"step": 760
|
| 578 |
},
|
| 579 |
{
|
| 580 |
"epoch": 4.028108108108108,
|
| 581 |
-
"grad_norm":
|
| 582 |
"learning_rate": 9.314903846153847e-06,
|
| 583 |
-
"loss": 0.
|
| 584 |
"step": 770
|
| 585 |
},
|
| 586 |
{
|
| 587 |
"epoch": 4.0389189189189185,
|
| 588 |
-
"grad_norm": 0.
|
| 589 |
"learning_rate": 8.713942307692307e-06,
|
| 590 |
-
"loss": 0.
|
| 591 |
"step": 780
|
| 592 |
},
|
| 593 |
{
|
| 594 |
"epoch": 4.04972972972973,
|
| 595 |
-
"grad_norm":
|
| 596 |
"learning_rate": 8.112980769230769e-06,
|
| 597 |
-
"loss": 0.
|
| 598 |
"step": 790
|
| 599 |
},
|
| 600 |
{
|
| 601 |
"epoch": 4.060540540540541,
|
| 602 |
-
"grad_norm": 0.
|
| 603 |
"learning_rate": 7.512019230769231e-06,
|
| 604 |
-
"loss": 0.
|
| 605 |
"step": 800
|
| 606 |
},
|
| 607 |
{
|
| 608 |
"epoch": 4.0713513513513515,
|
| 609 |
-
"grad_norm": 0.
|
| 610 |
"learning_rate": 6.911057692307693e-06,
|
| 611 |
-
"loss": 0.
|
| 612 |
"step": 810
|
| 613 |
},
|
| 614 |
{
|
| 615 |
"epoch": 4.082162162162162,
|
| 616 |
-
"grad_norm": 0.
|
| 617 |
"learning_rate": 6.310096153846154e-06,
|
| 618 |
-
"loss": 0.
|
| 619 |
"step": 820
|
| 620 |
},
|
| 621 |
{
|
| 622 |
"epoch": 4.092972972972973,
|
| 623 |
-
"grad_norm": 0.
|
| 624 |
"learning_rate": 5.709134615384616e-06,
|
| 625 |
-
"loss": 0.
|
| 626 |
"step": 830
|
| 627 |
},
|
| 628 |
{
|
| 629 |
"epoch": 4.103783783783784,
|
| 630 |
-
"grad_norm":
|
| 631 |
"learning_rate": 5.108173076923077e-06,
|
| 632 |
-
"loss": 0.
|
| 633 |
"step": 840
|
| 634 |
},
|
| 635 |
{
|
| 636 |
"epoch": 4.114594594594594,
|
| 637 |
-
"grad_norm": 0.
|
| 638 |
"learning_rate": 4.507211538461539e-06,
|
| 639 |
-
"loss": 0.
|
| 640 |
"step": 850
|
| 641 |
},
|
| 642 |
{
|
| 643 |
"epoch": 4.125405405405405,
|
| 644 |
-
"grad_norm":
|
| 645 |
"learning_rate": 3.90625e-06,
|
| 646 |
-
"loss": 0.
|
| 647 |
"step": 860
|
| 648 |
},
|
| 649 |
{
|
| 650 |
"epoch": 4.136216216216217,
|
| 651 |
-
"grad_norm": 0.
|
| 652 |
"learning_rate": 3.3052884615384617e-06,
|
| 653 |
-
"loss": 0.
|
| 654 |
"step": 870
|
| 655 |
},
|
| 656 |
{
|
| 657 |
"epoch": 4.147027027027027,
|
| 658 |
-
"grad_norm":
|
| 659 |
"learning_rate": 2.7043269230769233e-06,
|
| 660 |
-
"loss": 0.
|
| 661 |
"step": 880
|
| 662 |
},
|
| 663 |
{
|
| 664 |
"epoch": 4.157837837837838,
|
| 665 |
-
"grad_norm": 0.
|
| 666 |
"learning_rate": 2.103365384615385e-06,
|
| 667 |
-
"loss": 0.
|
| 668 |
"step": 890
|
| 669 |
},
|
| 670 |
{
|
| 671 |
"epoch": 4.168648648648649,
|
| 672 |
-
"grad_norm": 0.
|
| 673 |
"learning_rate": 1.5024038461538464e-06,
|
| 674 |
-
"loss": 0.
|
| 675 |
"step": 900
|
| 676 |
},
|
| 677 |
{
|
| 678 |
"epoch": 4.179459459459459,
|
| 679 |
-
"grad_norm":
|
| 680 |
"learning_rate": 9.014423076923077e-07,
|
| 681 |
-
"loss": 0.
|
| 682 |
"step": 910
|
| 683 |
},
|
| 684 |
{
|
| 685 |
"epoch": 4.19027027027027,
|
| 686 |
-
"grad_norm": 0.
|
| 687 |
"learning_rate": 3.0048076923076924e-07,
|
| 688 |
-
"loss": 0.
|
| 689 |
"step": 920
|
| 690 |
},
|
| 691 |
{
|
| 692 |
"epoch": 4.195675675675676,
|
| 693 |
-
"eval_accuracy": 0.
|
| 694 |
-
"eval_loss": 0.
|
| 695 |
-
"eval_runtime":
|
| 696 |
-
"eval_samples_per_second": 0.
|
| 697 |
-
"eval_steps_per_second": 0.
|
| 698 |
"step": 925
|
| 699 |
},
|
| 700 |
{
|
| 701 |
"epoch": 4.195675675675676,
|
| 702 |
"step": 925,
|
| 703 |
"total_flos": 3.2312722580082524e+18,
|
| 704 |
-
"train_loss": 0.
|
| 705 |
-
"train_runtime":
|
| 706 |
-
"train_samples_per_second": 0.
|
| 707 |
-
"train_steps_per_second": 0.
|
| 708 |
},
|
| 709 |
{
|
| 710 |
"epoch": 4.195675675675676,
|
| 711 |
-
"eval_accuracy": 0.
|
| 712 |
-
"eval_loss": 0.
|
| 713 |
-
"eval_runtime":
|
| 714 |
-
"eval_samples_per_second": 0.
|
| 715 |
-
"eval_steps_per_second": 0.
|
| 716 |
"step": 925
|
| 717 |
},
|
| 718 |
{
|
| 719 |
"epoch": 4.195675675675676,
|
| 720 |
-
"eval_accuracy": 0.
|
| 721 |
-
"eval_loss": 0.
|
| 722 |
-
"eval_runtime":
|
| 723 |
-
"eval_samples_per_second": 0.
|
| 724 |
-
"eval_steps_per_second": 0.
|
| 725 |
"step": 925
|
| 726 |
},
|
| 727 |
{
|
| 728 |
"epoch": 4.195675675675676,
|
| 729 |
-
"eval_accuracy": 0.
|
| 730 |
-
"eval_loss": 0.
|
| 731 |
-
"eval_runtime":
|
| 732 |
-
"eval_samples_per_second": 0.
|
| 733 |
-
"eval_steps_per_second": 0.
|
| 734 |
"step": 925
|
| 735 |
}
|
| 736 |
],
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.8381742738589212,
|
| 3 |
"best_model_checkpoint": "videomae-timesformer-surf-analytics\\checkpoint-925",
|
| 4 |
"epoch": 4.195675675675676,
|
| 5 |
"eval_steps": 500,
|
|
|
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.010810810810810811,
|
| 13 |
+
"grad_norm": 11.895417213439941,
|
| 14 |
"learning_rate": 5.376344086021506e-06,
|
| 15 |
+
"loss": 1.5009,
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.021621621621621623,
|
| 20 |
+
"grad_norm": 11.518338203430176,
|
| 21 |
"learning_rate": 1.0752688172043012e-05,
|
| 22 |
+
"loss": 1.1077,
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.032432432432432434,
|
| 27 |
+
"grad_norm": 10.138296127319336,
|
| 28 |
"learning_rate": 1.6129032258064517e-05,
|
| 29 |
+
"loss": 1.3124,
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.043243243243243246,
|
| 34 |
+
"grad_norm": 9.477767944335938,
|
| 35 |
"learning_rate": 2.1505376344086024e-05,
|
| 36 |
+
"loss": 1.0199,
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.05405405405405406,
|
| 41 |
+
"grad_norm": 11.046083450317383,
|
| 42 |
"learning_rate": 2.6881720430107527e-05,
|
| 43 |
+
"loss": 1.038,
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.06486486486486487,
|
| 48 |
+
"grad_norm": 12.39089584350586,
|
| 49 |
"learning_rate": 3.2258064516129034e-05,
|
| 50 |
+
"loss": 0.9406,
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.07567567567567568,
|
| 55 |
+
"grad_norm": 12.075498580932617,
|
| 56 |
"learning_rate": 3.763440860215054e-05,
|
| 57 |
+
"loss": 1.0054,
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.08648648648648649,
|
| 62 |
+
"grad_norm": 9.824885368347168,
|
| 63 |
"learning_rate": 4.301075268817205e-05,
|
| 64 |
+
"loss": 0.7344,
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.0972972972972973,
|
| 69 |
+
"grad_norm": 10.267470359802246,
|
| 70 |
"learning_rate": 4.8387096774193554e-05,
|
| 71 |
+
"loss": 0.6901,
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 0.10810810810810811,
|
| 76 |
+
"grad_norm": 4.5673370361328125,
|
| 77 |
"learning_rate": 4.957932692307692e-05,
|
| 78 |
+
"loss": 0.4999,
|
| 79 |
"step": 100
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 0.11891891891891893,
|
| 83 |
+
"grad_norm": 7.499106407165527,
|
| 84 |
"learning_rate": 4.897836538461539e-05,
|
| 85 |
+
"loss": 0.7291,
|
| 86 |
"step": 110
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 0.12972972972972974,
|
| 90 |
+
"grad_norm": 13.209514617919922,
|
| 91 |
"learning_rate": 4.8377403846153846e-05,
|
| 92 |
+
"loss": 0.4908,
|
| 93 |
"step": 120
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 0.14054054054054055,
|
| 97 |
+
"grad_norm": 27.258346557617188,
|
| 98 |
"learning_rate": 4.777644230769231e-05,
|
| 99 |
+
"loss": 0.6281,
|
| 100 |
"step": 130
|
| 101 |
},
|
| 102 |
{
|
| 103 |
"epoch": 0.15135135135135136,
|
| 104 |
+
"grad_norm": 23.13751220703125,
|
| 105 |
"learning_rate": 4.717548076923077e-05,
|
| 106 |
+
"loss": 0.7854,
|
| 107 |
"step": 140
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 0.16216216216216217,
|
| 111 |
+
"grad_norm": 21.831802368164062,
|
| 112 |
"learning_rate": 4.657451923076923e-05,
|
| 113 |
+
"loss": 0.6653,
|
| 114 |
"step": 150
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 0.17297297297297298,
|
| 118 |
+
"grad_norm": 3.0441360473632812,
|
| 119 |
"learning_rate": 4.5973557692307694e-05,
|
| 120 |
+
"loss": 1.2533,
|
| 121 |
"step": 160
|
| 122 |
},
|
| 123 |
{
|
| 124 |
"epoch": 0.1837837837837838,
|
| 125 |
+
"grad_norm": 10.554399490356445,
|
| 126 |
"learning_rate": 4.5372596153846156e-05,
|
| 127 |
+
"loss": 0.546,
|
| 128 |
"step": 170
|
| 129 |
},
|
| 130 |
{
|
| 131 |
"epoch": 0.1945945945945946,
|
| 132 |
+
"grad_norm": 14.284193992614746,
|
| 133 |
"learning_rate": 4.477163461538462e-05,
|
| 134 |
+
"loss": 0.4491,
|
| 135 |
"step": 180
|
| 136 |
},
|
| 137 |
{
|
| 138 |
"epoch": 0.20108108108108108,
|
| 139 |
+
"eval_accuracy": 0.7385892116182573,
|
| 140 |
+
"eval_loss": 0.6939307451248169,
|
| 141 |
+
"eval_runtime": 407.9813,
|
| 142 |
+
"eval_samples_per_second": 0.591,
|
| 143 |
+
"eval_steps_per_second": 0.15,
|
| 144 |
"step": 186
|
| 145 |
},
|
| 146 |
{
|
| 147 |
"epoch": 1.0043243243243243,
|
| 148 |
+
"grad_norm": 11.024110794067383,
|
| 149 |
"learning_rate": 4.417067307692308e-05,
|
| 150 |
+
"loss": 0.4377,
|
| 151 |
"step": 190
|
| 152 |
},
|
| 153 |
{
|
| 154 |
"epoch": 1.0151351351351352,
|
| 155 |
+
"grad_norm": 4.518445014953613,
|
| 156 |
"learning_rate": 4.3569711538461535e-05,
|
| 157 |
+
"loss": 0.0803,
|
| 158 |
"step": 200
|
| 159 |
},
|
| 160 |
{
|
| 161 |
"epoch": 1.025945945945946,
|
| 162 |
+
"grad_norm": 0.42581477761268616,
|
| 163 |
"learning_rate": 4.2968750000000004e-05,
|
| 164 |
+
"loss": 0.3532,
|
| 165 |
"step": 210
|
| 166 |
},
|
| 167 |
{
|
| 168 |
"epoch": 1.0367567567567568,
|
| 169 |
+
"grad_norm": 13.55870246887207,
|
| 170 |
"learning_rate": 4.2367788461538466e-05,
|
| 171 |
+
"loss": 0.7917,
|
| 172 |
"step": 220
|
| 173 |
},
|
| 174 |
{
|
| 175 |
"epoch": 1.0475675675675675,
|
| 176 |
+
"grad_norm": 13.94046401977539,
|
| 177 |
"learning_rate": 4.176682692307692e-05,
|
| 178 |
+
"loss": 0.1946,
|
| 179 |
"step": 230
|
| 180 |
},
|
| 181 |
{
|
| 182 |
"epoch": 1.0583783783783784,
|
| 183 |
+
"grad_norm": 24.760360717773438,
|
| 184 |
"learning_rate": 4.116586538461539e-05,
|
| 185 |
+
"loss": 0.9298,
|
| 186 |
"step": 240
|
| 187 |
},
|
| 188 |
{
|
| 189 |
"epoch": 1.0691891891891891,
|
| 190 |
+
"grad_norm": 7.651761531829834,
|
| 191 |
"learning_rate": 4.0564903846153846e-05,
|
| 192 |
+
"loss": 0.7205,
|
| 193 |
"step": 250
|
| 194 |
},
|
| 195 |
{
|
| 196 |
"epoch": 1.08,
|
| 197 |
+
"grad_norm": 16.21826171875,
|
| 198 |
"learning_rate": 3.996394230769231e-05,
|
| 199 |
+
"loss": 0.6958,
|
| 200 |
"step": 260
|
| 201 |
},
|
| 202 |
{
|
| 203 |
"epoch": 1.0908108108108108,
|
| 204 |
+
"grad_norm": 16.531909942626953,
|
| 205 |
"learning_rate": 3.936298076923077e-05,
|
| 206 |
+
"loss": 0.2937,
|
| 207 |
"step": 270
|
| 208 |
},
|
| 209 |
{
|
| 210 |
"epoch": 1.1016216216216217,
|
| 211 |
+
"grad_norm": 16.31216049194336,
|
| 212 |
"learning_rate": 3.876201923076923e-05,
|
| 213 |
+
"loss": 0.6104,
|
| 214 |
"step": 280
|
| 215 |
},
|
| 216 |
{
|
| 217 |
"epoch": 1.1124324324324324,
|
| 218 |
+
"grad_norm": 8.48189640045166,
|
| 219 |
"learning_rate": 3.8161057692307694e-05,
|
| 220 |
+
"loss": 0.4254,
|
| 221 |
"step": 290
|
| 222 |
},
|
| 223 |
{
|
| 224 |
"epoch": 1.1232432432432433,
|
| 225 |
+
"grad_norm": 29.133272171020508,
|
| 226 |
"learning_rate": 3.7560096153846156e-05,
|
| 227 |
+
"loss": 0.262,
|
| 228 |
"step": 300
|
| 229 |
},
|
| 230 |
{
|
| 231 |
"epoch": 1.134054054054054,
|
| 232 |
+
"grad_norm": 20.123014450073242,
|
| 233 |
"learning_rate": 3.695913461538462e-05,
|
| 234 |
+
"loss": 0.4921,
|
| 235 |
"step": 310
|
| 236 |
},
|
| 237 |
{
|
| 238 |
"epoch": 1.144864864864865,
|
| 239 |
+
"grad_norm": 9.895979881286621,
|
| 240 |
"learning_rate": 3.635817307692308e-05,
|
| 241 |
+
"loss": 0.2059,
|
| 242 |
"step": 320
|
| 243 |
},
|
| 244 |
{
|
| 245 |
"epoch": 1.1556756756756756,
|
| 246 |
+
"grad_norm": 11.974223136901855,
|
| 247 |
"learning_rate": 3.5757211538461535e-05,
|
| 248 |
+
"loss": 0.2591,
|
| 249 |
"step": 330
|
| 250 |
},
|
| 251 |
{
|
| 252 |
"epoch": 1.1664864864864866,
|
| 253 |
+
"grad_norm": 0.1094137504696846,
|
| 254 |
"learning_rate": 3.5156250000000004e-05,
|
| 255 |
+
"loss": 0.437,
|
| 256 |
"step": 340
|
| 257 |
},
|
| 258 |
{
|
| 259 |
"epoch": 1.1772972972972973,
|
| 260 |
+
"grad_norm": 0.12011051177978516,
|
| 261 |
"learning_rate": 3.4555288461538466e-05,
|
| 262 |
+
"loss": 0.5299,
|
| 263 |
"step": 350
|
| 264 |
},
|
| 265 |
{
|
| 266 |
"epoch": 1.1881081081081082,
|
| 267 |
+
"grad_norm": 0.32698747515678406,
|
| 268 |
"learning_rate": 3.395432692307692e-05,
|
| 269 |
+
"loss": 0.1667,
|
| 270 |
"step": 360
|
| 271 |
},
|
| 272 |
{
|
| 273 |
"epoch": 1.1989189189189189,
|
| 274 |
+
"grad_norm": 2.423952102661133,
|
| 275 |
"learning_rate": 3.335336538461539e-05,
|
| 276 |
+
"loss": 0.5627,
|
| 277 |
"step": 370
|
| 278 |
},
|
| 279 |
{
|
| 280 |
"epoch": 1.201081081081081,
|
| 281 |
+
"eval_accuracy": 0.7759336099585062,
|
| 282 |
+
"eval_loss": 0.6805610060691833,
|
| 283 |
+
"eval_runtime": 399.3517,
|
| 284 |
+
"eval_samples_per_second": 0.603,
|
| 285 |
+
"eval_steps_per_second": 0.153,
|
| 286 |
"step": 372
|
| 287 |
},
|
| 288 |
{
|
| 289 |
"epoch": 2.0086486486486486,
|
| 290 |
+
"grad_norm": 0.037202008068561554,
|
| 291 |
"learning_rate": 3.2752403846153846e-05,
|
| 292 |
+
"loss": 0.0415,
|
| 293 |
"step": 380
|
| 294 |
},
|
| 295 |
{
|
| 296 |
"epoch": 2.0194594594594593,
|
| 297 |
+
"grad_norm": 23.98035430908203,
|
| 298 |
"learning_rate": 3.215144230769231e-05,
|
| 299 |
+
"loss": 0.3379,
|
| 300 |
"step": 390
|
| 301 |
},
|
| 302 |
{
|
| 303 |
"epoch": 2.0302702702702704,
|
| 304 |
+
"grad_norm": 31.957252502441406,
|
| 305 |
"learning_rate": 3.155048076923077e-05,
|
| 306 |
+
"loss": 0.357,
|
| 307 |
"step": 400
|
| 308 |
},
|
| 309 |
{
|
| 310 |
"epoch": 2.041081081081081,
|
| 311 |
+
"grad_norm": 1.1118696928024292,
|
| 312 |
"learning_rate": 3.094951923076923e-05,
|
| 313 |
+
"loss": 0.1775,
|
| 314 |
"step": 410
|
| 315 |
},
|
| 316 |
{
|
| 317 |
"epoch": 2.051891891891892,
|
| 318 |
+
"grad_norm": 0.05329513177275658,
|
| 319 |
"learning_rate": 3.0348557692307694e-05,
|
| 320 |
+
"loss": 0.0175,
|
| 321 |
"step": 420
|
| 322 |
},
|
| 323 |
{
|
| 324 |
"epoch": 2.0627027027027025,
|
| 325 |
+
"grad_norm": 32.40390396118164,
|
| 326 |
"learning_rate": 2.974759615384616e-05,
|
| 327 |
+
"loss": 0.4569,
|
| 328 |
"step": 430
|
| 329 |
},
|
| 330 |
{
|
| 331 |
"epoch": 2.0735135135135137,
|
| 332 |
+
"grad_norm": 0.15056586265563965,
|
| 333 |
"learning_rate": 2.9146634615384614e-05,
|
| 334 |
+
"loss": 0.3911,
|
| 335 |
"step": 440
|
| 336 |
},
|
| 337 |
{
|
| 338 |
"epoch": 2.0843243243243244,
|
| 339 |
+
"grad_norm": 0.2210923731327057,
|
| 340 |
"learning_rate": 2.854567307692308e-05,
|
| 341 |
+
"loss": 0.112,
|
| 342 |
"step": 450
|
| 343 |
},
|
| 344 |
{
|
| 345 |
"epoch": 2.095135135135135,
|
| 346 |
+
"grad_norm": 0.9941416382789612,
|
| 347 |
"learning_rate": 2.794471153846154e-05,
|
| 348 |
+
"loss": 0.3037,
|
| 349 |
"step": 460
|
| 350 |
},
|
| 351 |
{
|
| 352 |
"epoch": 2.1059459459459458,
|
| 353 |
+
"grad_norm": 0.035190433263778687,
|
| 354 |
"learning_rate": 2.734375e-05,
|
| 355 |
+
"loss": 0.6371,
|
| 356 |
"step": 470
|
| 357 |
},
|
| 358 |
{
|
| 359 |
"epoch": 2.116756756756757,
|
| 360 |
+
"grad_norm": 0.04186534136533737,
|
| 361 |
"learning_rate": 2.6742788461538466e-05,
|
| 362 |
+
"loss": 0.4244,
|
| 363 |
"step": 480
|
| 364 |
},
|
| 365 |
{
|
| 366 |
"epoch": 2.1275675675675676,
|
| 367 |
+
"grad_norm": 0.5409103631973267,
|
| 368 |
"learning_rate": 2.6141826923076925e-05,
|
| 369 |
+
"loss": 0.311,
|
| 370 |
"step": 490
|
| 371 |
},
|
| 372 |
{
|
| 373 |
"epoch": 2.1383783783783783,
|
| 374 |
+
"grad_norm": 0.04897398501634598,
|
| 375 |
"learning_rate": 2.5540865384615387e-05,
|
| 376 |
+
"loss": 0.0555,
|
| 377 |
"step": 500
|
| 378 |
},
|
| 379 |
{
|
| 380 |
"epoch": 2.149189189189189,
|
| 381 |
+
"grad_norm": 16.756502151489258,
|
| 382 |
"learning_rate": 2.493990384615385e-05,
|
| 383 |
+
"loss": 0.264,
|
| 384 |
"step": 510
|
| 385 |
},
|
| 386 |
{
|
| 387 |
"epoch": 2.16,
|
| 388 |
+
"grad_norm": 1.5812554359436035,
|
| 389 |
"learning_rate": 2.4338942307692307e-05,
|
| 390 |
+
"loss": 0.1869,
|
| 391 |
"step": 520
|
| 392 |
},
|
| 393 |
{
|
| 394 |
"epoch": 2.170810810810811,
|
| 395 |
+
"grad_norm": 0.20836937427520752,
|
| 396 |
"learning_rate": 2.373798076923077e-05,
|
| 397 |
+
"loss": 0.2804,
|
| 398 |
"step": 530
|
| 399 |
},
|
| 400 |
{
|
| 401 |
"epoch": 2.1816216216216215,
|
| 402 |
+
"grad_norm": 0.14942559599876404,
|
| 403 |
"learning_rate": 2.313701923076923e-05,
|
| 404 |
+
"loss": 0.2552,
|
| 405 |
"step": 540
|
| 406 |
},
|
| 407 |
{
|
| 408 |
"epoch": 2.1924324324324322,
|
| 409 |
+
"grad_norm": 0.09908430278301239,
|
| 410 |
"learning_rate": 2.2536057692307694e-05,
|
| 411 |
+
"loss": 0.5189,
|
| 412 |
"step": 550
|
| 413 |
},
|
| 414 |
{
|
| 415 |
"epoch": 2.2010810810810812,
|
| 416 |
+
"eval_accuracy": 0.8174273858921162,
|
| 417 |
+
"eval_loss": 0.651021420955658,
|
| 418 |
+
"eval_runtime": 406.013,
|
| 419 |
+
"eval_samples_per_second": 0.594,
|
| 420 |
+
"eval_steps_per_second": 0.15,
|
| 421 |
"step": 558
|
| 422 |
},
|
| 423 |
{
|
| 424 |
"epoch": 3.002162162162162,
|
| 425 |
+
"grad_norm": 15.29819107055664,
|
| 426 |
"learning_rate": 2.1935096153846156e-05,
|
| 427 |
+
"loss": 0.7952,
|
| 428 |
"step": 560
|
| 429 |
},
|
| 430 |
{
|
| 431 |
"epoch": 3.012972972972973,
|
| 432 |
+
"grad_norm": 7.53931188583374,
|
| 433 |
"learning_rate": 2.1334134615384614e-05,
|
| 434 |
+
"loss": 0.0482,
|
| 435 |
"step": 570
|
| 436 |
},
|
| 437 |
{
|
| 438 |
"epoch": 3.023783783783784,
|
| 439 |
+
"grad_norm": 0.01849238947033882,
|
| 440 |
"learning_rate": 2.073317307692308e-05,
|
| 441 |
+
"loss": 0.0809,
|
| 442 |
"step": 580
|
| 443 |
},
|
| 444 |
{
|
| 445 |
"epoch": 3.0345945945945947,
|
| 446 |
+
"grad_norm": 28.94001007080078,
|
| 447 |
"learning_rate": 2.0132211538461542e-05,
|
| 448 |
+
"loss": 0.0708,
|
| 449 |
"step": 590
|
| 450 |
},
|
| 451 |
{
|
| 452 |
"epoch": 3.0454054054054054,
|
| 453 |
+
"grad_norm": 0.031998779624700546,
|
| 454 |
"learning_rate": 1.953125e-05,
|
| 455 |
+
"loss": 0.0426,
|
| 456 |
"step": 600
|
| 457 |
},
|
| 458 |
{
|
| 459 |
"epoch": 3.056216216216216,
|
| 460 |
+
"grad_norm": 6.872684001922607,
|
| 461 |
"learning_rate": 1.8930288461538462e-05,
|
| 462 |
+
"loss": 0.1062,
|
| 463 |
"step": 610
|
| 464 |
},
|
| 465 |
{
|
| 466 |
"epoch": 3.0670270270270272,
|
| 467 |
+
"grad_norm": 0.02178012765944004,
|
| 468 |
"learning_rate": 1.832932692307692e-05,
|
| 469 |
+
"loss": 0.1908,
|
| 470 |
"step": 620
|
| 471 |
},
|
| 472 |
{
|
| 473 |
"epoch": 3.077837837837838,
|
| 474 |
+
"grad_norm": 12.06849193572998,
|
| 475 |
"learning_rate": 1.7728365384615387e-05,
|
| 476 |
+
"loss": 0.0114,
|
| 477 |
"step": 630
|
| 478 |
},
|
| 479 |
{
|
| 480 |
"epoch": 3.0886486486486486,
|
| 481 |
+
"grad_norm": 0.18928392231464386,
|
| 482 |
"learning_rate": 1.712740384615385e-05,
|
| 483 |
+
"loss": 0.1763,
|
| 484 |
"step": 640
|
| 485 |
},
|
| 486 |
{
|
| 487 |
"epoch": 3.0994594594594593,
|
| 488 |
+
"grad_norm": 10.532954216003418,
|
| 489 |
"learning_rate": 1.6526442307692307e-05,
|
| 490 |
+
"loss": 0.3634,
|
| 491 |
"step": 650
|
| 492 |
},
|
| 493 |
{
|
| 494 |
"epoch": 3.1102702702702705,
|
| 495 |
+
"grad_norm": 18.90543556213379,
|
| 496 |
"learning_rate": 1.592548076923077e-05,
|
| 497 |
+
"loss": 0.728,
|
| 498 |
"step": 660
|
| 499 |
},
|
| 500 |
{
|
| 501 |
"epoch": 3.121081081081081,
|
| 502 |
+
"grad_norm": 0.35980573296546936,
|
| 503 |
"learning_rate": 1.532451923076923e-05,
|
| 504 |
+
"loss": 0.144,
|
| 505 |
"step": 670
|
| 506 |
},
|
| 507 |
{
|
| 508 |
"epoch": 3.131891891891892,
|
| 509 |
+
"grad_norm": 0.03495261073112488,
|
| 510 |
"learning_rate": 1.4723557692307693e-05,
|
| 511 |
+
"loss": 0.0163,
|
| 512 |
"step": 680
|
| 513 |
},
|
| 514 |
{
|
| 515 |
"epoch": 3.1427027027027026,
|
| 516 |
+
"grad_norm": 24.22405242919922,
|
| 517 |
"learning_rate": 1.4122596153846154e-05,
|
| 518 |
+
"loss": 0.1584,
|
| 519 |
"step": 690
|
| 520 |
},
|
| 521 |
{
|
| 522 |
"epoch": 3.1535135135135137,
|
| 523 |
+
"grad_norm": 0.026546325534582138,
|
| 524 |
"learning_rate": 1.3521634615384616e-05,
|
| 525 |
+
"loss": 0.3465,
|
| 526 |
"step": 700
|
| 527 |
},
|
| 528 |
{
|
| 529 |
"epoch": 3.1643243243243244,
|
| 530 |
+
"grad_norm": 0.43370339274406433,
|
| 531 |
"learning_rate": 1.292067307692308e-05,
|
| 532 |
+
"loss": 0.3739,
|
| 533 |
"step": 710
|
| 534 |
},
|
| 535 |
{
|
| 536 |
"epoch": 3.175135135135135,
|
| 537 |
+
"grad_norm": 0.02699950709939003,
|
| 538 |
"learning_rate": 1.231971153846154e-05,
|
| 539 |
+
"loss": 0.1644,
|
| 540 |
"step": 720
|
| 541 |
},
|
| 542 |
{
|
| 543 |
"epoch": 3.185945945945946,
|
| 544 |
+
"grad_norm": 2.032027006149292,
|
| 545 |
"learning_rate": 1.171875e-05,
|
| 546 |
+
"loss": 0.1225,
|
| 547 |
"step": 730
|
| 548 |
},
|
| 549 |
{
|
| 550 |
"epoch": 3.1967567567567565,
|
| 551 |
+
"grad_norm": 0.006064319983124733,
|
| 552 |
"learning_rate": 1.111778846153846e-05,
|
| 553 |
+
"loss": 0.2503,
|
| 554 |
"step": 740
|
| 555 |
},
|
| 556 |
{
|
| 557 |
"epoch": 3.2010810810810812,
|
| 558 |
+
"eval_accuracy": 0.8174273858921162,
|
| 559 |
+
"eval_loss": 0.6731572151184082,
|
| 560 |
+
"eval_runtime": 400.1385,
|
| 561 |
+
"eval_samples_per_second": 0.602,
|
| 562 |
+
"eval_steps_per_second": 0.152,
|
| 563 |
"step": 744
|
| 564 |
},
|
| 565 |
{
|
| 566 |
"epoch": 4.006486486486486,
|
| 567 |
+
"grad_norm": 0.4040788412094116,
|
| 568 |
"learning_rate": 1.0516826923076924e-05,
|
| 569 |
+
"loss": 0.2375,
|
| 570 |
"step": 750
|
| 571 |
},
|
| 572 |
{
|
| 573 |
"epoch": 4.017297297297297,
|
| 574 |
+
"grad_norm": 11.23088264465332,
|
| 575 |
"learning_rate": 9.915865384615385e-06,
|
| 576 |
+
"loss": 0.1009,
|
| 577 |
"step": 760
|
| 578 |
},
|
| 579 |
{
|
| 580 |
"epoch": 4.028108108108108,
|
| 581 |
+
"grad_norm": 44.4477424621582,
|
| 582 |
"learning_rate": 9.314903846153847e-06,
|
| 583 |
+
"loss": 0.1268,
|
| 584 |
"step": 770
|
| 585 |
},
|
| 586 |
{
|
| 587 |
"epoch": 4.0389189189189185,
|
| 588 |
+
"grad_norm": 0.01056050043553114,
|
| 589 |
"learning_rate": 8.713942307692307e-06,
|
| 590 |
+
"loss": 0.0738,
|
| 591 |
"step": 780
|
| 592 |
},
|
| 593 |
{
|
| 594 |
"epoch": 4.04972972972973,
|
| 595 |
+
"grad_norm": 14.111398696899414,
|
| 596 |
"learning_rate": 8.112980769230769e-06,
|
| 597 |
+
"loss": 0.2126,
|
| 598 |
"step": 790
|
| 599 |
},
|
| 600 |
{
|
| 601 |
"epoch": 4.060540540540541,
|
| 602 |
+
"grad_norm": 0.014713864773511887,
|
| 603 |
"learning_rate": 7.512019230769231e-06,
|
| 604 |
+
"loss": 0.0011,
|
| 605 |
"step": 800
|
| 606 |
},
|
| 607 |
{
|
| 608 |
"epoch": 4.0713513513513515,
|
| 609 |
+
"grad_norm": 0.010851687751710415,
|
| 610 |
"learning_rate": 6.911057692307693e-06,
|
| 611 |
+
"loss": 0.1786,
|
| 612 |
"step": 810
|
| 613 |
},
|
| 614 |
{
|
| 615 |
"epoch": 4.082162162162162,
|
| 616 |
+
"grad_norm": 0.050808586180210114,
|
| 617 |
"learning_rate": 6.310096153846154e-06,
|
| 618 |
+
"loss": 0.0783,
|
| 619 |
"step": 820
|
| 620 |
},
|
| 621 |
{
|
| 622 |
"epoch": 4.092972972972973,
|
| 623 |
+
"grad_norm": 0.037915121763944626,
|
| 624 |
"learning_rate": 5.709134615384616e-06,
|
| 625 |
+
"loss": 0.2292,
|
| 626 |
"step": 830
|
| 627 |
},
|
| 628 |
{
|
| 629 |
"epoch": 4.103783783783784,
|
| 630 |
+
"grad_norm": 2.1745355129241943,
|
| 631 |
"learning_rate": 5.108173076923077e-06,
|
| 632 |
+
"loss": 0.1868,
|
| 633 |
"step": 840
|
| 634 |
},
|
| 635 |
{
|
| 636 |
"epoch": 4.114594594594594,
|
| 637 |
+
"grad_norm": 0.020182184875011444,
|
| 638 |
"learning_rate": 4.507211538461539e-06,
|
| 639 |
+
"loss": 0.0016,
|
| 640 |
"step": 850
|
| 641 |
},
|
| 642 |
{
|
| 643 |
"epoch": 4.125405405405405,
|
| 644 |
+
"grad_norm": 23.419614791870117,
|
| 645 |
"learning_rate": 3.90625e-06,
|
| 646 |
+
"loss": 0.0688,
|
| 647 |
"step": 860
|
| 648 |
},
|
| 649 |
{
|
| 650 |
"epoch": 4.136216216216217,
|
| 651 |
+
"grad_norm": 0.5133500695228577,
|
| 652 |
"learning_rate": 3.3052884615384617e-06,
|
| 653 |
+
"loss": 0.0782,
|
| 654 |
"step": 870
|
| 655 |
},
|
| 656 |
{
|
| 657 |
"epoch": 4.147027027027027,
|
| 658 |
+
"grad_norm": 1.0855499505996704,
|
| 659 |
"learning_rate": 2.7043269230769233e-06,
|
| 660 |
+
"loss": 0.0486,
|
| 661 |
"step": 880
|
| 662 |
},
|
| 663 |
{
|
| 664 |
"epoch": 4.157837837837838,
|
| 665 |
+
"grad_norm": 0.05244865640997887,
|
| 666 |
"learning_rate": 2.103365384615385e-06,
|
| 667 |
+
"loss": 0.0217,
|
| 668 |
"step": 890
|
| 669 |
},
|
| 670 |
{
|
| 671 |
"epoch": 4.168648648648649,
|
| 672 |
+
"grad_norm": 0.013233544304966927,
|
| 673 |
"learning_rate": 1.5024038461538464e-06,
|
| 674 |
+
"loss": 0.2434,
|
| 675 |
"step": 900
|
| 676 |
},
|
| 677 |
{
|
| 678 |
"epoch": 4.179459459459459,
|
| 679 |
+
"grad_norm": 12.086763381958008,
|
| 680 |
"learning_rate": 9.014423076923077e-07,
|
| 681 |
+
"loss": 0.0501,
|
| 682 |
"step": 910
|
| 683 |
},
|
| 684 |
{
|
| 685 |
"epoch": 4.19027027027027,
|
| 686 |
+
"grad_norm": 0.0869203507900238,
|
| 687 |
"learning_rate": 3.0048076923076924e-07,
|
| 688 |
+
"loss": 0.0159,
|
| 689 |
"step": 920
|
| 690 |
},
|
| 691 |
{
|
| 692 |
"epoch": 4.195675675675676,
|
| 693 |
+
"eval_accuracy": 0.8381742738589212,
|
| 694 |
+
"eval_loss": 0.6192311644554138,
|
| 695 |
+
"eval_runtime": 401.9033,
|
| 696 |
+
"eval_samples_per_second": 0.6,
|
| 697 |
+
"eval_steps_per_second": 0.152,
|
| 698 |
"step": 925
|
| 699 |
},
|
| 700 |
{
|
| 701 |
"epoch": 4.195675675675676,
|
| 702 |
"step": 925,
|
| 703 |
"total_flos": 3.2312722580082524e+18,
|
| 704 |
+
"train_loss": 0.3796664474924674,
|
| 705 |
+
"train_runtime": 18773.5619,
|
| 706 |
+
"train_samples_per_second": 0.197,
|
| 707 |
+
"train_steps_per_second": 0.049
|
| 708 |
},
|
| 709 |
{
|
| 710 |
"epoch": 4.195675675675676,
|
| 711 |
+
"eval_accuracy": 0.9770580296896086,
|
| 712 |
+
"eval_loss": 0.11411414295434952,
|
| 713 |
+
"eval_runtime": 1291.965,
|
| 714 |
+
"eval_samples_per_second": 0.574,
|
| 715 |
+
"eval_steps_per_second": 0.144,
|
| 716 |
"step": 925
|
| 717 |
},
|
| 718 |
{
|
| 719 |
"epoch": 4.195675675675676,
|
| 720 |
+
"eval_accuracy": 0.8134831460674158,
|
| 721 |
+
"eval_loss": 0.7141955494880676,
|
| 722 |
+
"eval_runtime": 720.0647,
|
| 723 |
+
"eval_samples_per_second": 0.618,
|
| 724 |
+
"eval_steps_per_second": 0.156,
|
| 725 |
"step": 925
|
| 726 |
},
|
| 727 |
{
|
| 728 |
"epoch": 4.195675675675676,
|
| 729 |
+
"eval_accuracy": 0.8381742738589212,
|
| 730 |
+
"eval_loss": 0.6192311644554138,
|
| 731 |
+
"eval_runtime": 391.274,
|
| 732 |
+
"eval_samples_per_second": 0.616,
|
| 733 |
+
"eval_steps_per_second": 0.156,
|
| 734 |
"step": 925
|
| 735 |
}
|
| 736 |
],
|
val_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 4.195675675675676,
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second": 0.
|
| 7 |
-
"eval_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 4.195675675675676,
|
| 3 |
+
"eval_accuracy": 0.8381742738589212,
|
| 4 |
+
"eval_loss": 0.6192311644554138,
|
| 5 |
+
"eval_runtime": 391.274,
|
| 6 |
+
"eval_samples_per_second": 0.616,
|
| 7 |
+
"eval_steps_per_second": 0.156
|
| 8 |
}
|