dhritic99/model99123
Browse files- README.md +10 -10
- all_results.json +9 -9
- eval_results.json +5 -5
- model.safetensors +1 -1
- train_results.json +4 -4
- trainer_state.json +269 -269
- training_args.bin +1 -1
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 [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) 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 |
|
|
@@ -37,7 +37,7 @@ More information needed
|
|
| 37 |
### Training hyperparameters
|
| 38 |
|
| 39 |
The following hyperparameters were used during training:
|
| 40 |
-
- learning_rate:
|
| 41 |
- train_batch_size: 16
|
| 42 |
- eval_batch_size: 8
|
| 43 |
- seed: 42
|
|
@@ -53,13 +53,13 @@ The following hyperparameters were used during training:
|
|
| 53 |
|
| 54 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 55 |
|:-------------:|:-------:|:----:|:---------------:|:--------:|
|
| 56 |
-
| 0.
|
| 57 |
-
| 0.
|
| 58 |
-
| 0.
|
| 59 |
-
| 0.
|
| 60 |
-
| 0.
|
| 61 |
-
| 0.
|
| 62 |
-
| 0.
|
| 63 |
|
| 64 |
|
| 65 |
### Framework versions
|
|
|
|
| 17 |
|
| 18 |
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on an unknown dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 0.3419
|
| 21 |
+
- Accuracy: 0.9406
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
|
|
| 37 |
### Training hyperparameters
|
| 38 |
|
| 39 |
The following hyperparameters were used during training:
|
| 40 |
+
- learning_rate: 1e-05
|
| 41 |
- train_batch_size: 16
|
| 42 |
- eval_batch_size: 8
|
| 43 |
- seed: 42
|
|
|
|
| 53 |
|
| 54 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 55 |
|:-------------:|:-------:|:----:|:---------------:|:--------:|
|
| 56 |
+
| 0.0048 | 7.8125 | 500 | 0.2337 | 0.9473 |
|
| 57 |
+
| 0.0012 | 15.625 | 1000 | 0.1950 | 0.9531 |
|
| 58 |
+
| 0.0007 | 23.4375 | 1500 | 0.1927 | 0.9580 |
|
| 59 |
+
| 0.0004 | 31.25 | 2000 | 0.1970 | 0.9629 |
|
| 60 |
+
| 0.0003 | 39.0625 | 2500 | 0.2040 | 0.9629 |
|
| 61 |
+
| 0.0002 | 46.875 | 3000 | 0.2114 | 0.9629 |
|
| 62 |
+
| 0.0002 | 54.6875 | 3500 | 0.2171 | 0.9648 |
|
| 63 |
|
| 64 |
|
| 65 |
### Framework versions
|
all_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 60.0,
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime": 6.
|
| 6 |
-
"eval_samples_per_second": 190.
|
| 7 |
-
"eval_steps_per_second": 23.
|
| 8 |
"total_flos": 1.904477274611122e+19,
|
| 9 |
-
"train_loss": 0.
|
| 10 |
-
"train_runtime":
|
| 11 |
-
"train_samples_per_second":
|
| 12 |
-
"train_steps_per_second": 1.
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 60.0,
|
| 3 |
+
"eval_accuracy": 0.940625,
|
| 4 |
+
"eval_loss": 0.3418883681297302,
|
| 5 |
+
"eval_runtime": 6.7246,
|
| 6 |
+
"eval_samples_per_second": 190.345,
|
| 7 |
+
"eval_steps_per_second": 23.793,
|
| 8 |
"total_flos": 1.904477274611122e+19,
|
| 9 |
+
"train_loss": 0.0010260362852325974,
|
| 10 |
+
"train_runtime": 2806.5312,
|
| 11 |
+
"train_samples_per_second": 87.567,
|
| 12 |
+
"train_steps_per_second": 1.368
|
| 13 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 60.0,
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime": 6.
|
| 6 |
-
"eval_samples_per_second": 190.
|
| 7 |
-
"eval_steps_per_second": 23.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 60.0,
|
| 3 |
+
"eval_accuracy": 0.940625,
|
| 4 |
+
"eval_loss": 0.3418883681297302,
|
| 5 |
+
"eval_runtime": 6.7246,
|
| 6 |
+
"eval_samples_per_second": 190.345,
|
| 7 |
+
"eval_steps_per_second": 23.793
|
| 8 |
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 343230128
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cd9f5e271ad1d8119de83579c28db38d081d9cf64bf984cb94bcef6fec2881b1
|
| 3 |
size 343230128
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 60.0,
|
| 3 |
"total_flos": 1.904477274611122e+19,
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 1.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 60.0,
|
| 3 |
"total_flos": 1.904477274611122e+19,
|
| 4 |
+
"train_loss": 0.0010260362852325974,
|
| 5 |
+
"train_runtime": 2806.5312,
|
| 6 |
+
"train_samples_per_second": 87.567,
|
| 7 |
+
"train_steps_per_second": 1.368
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "./vit-base-brain-tumor-detection3/checkpoint-
|
| 4 |
"epoch": 60.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
"global_step": 3840,
|
|
@@ -10,607 +10,607 @@
|
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.78125,
|
| 13 |
-
"grad_norm":
|
| 14 |
-
"learning_rate":
|
| 15 |
-
"loss":
|
| 16 |
"step": 50
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 1.5625,
|
| 20 |
-
"grad_norm":
|
| 21 |
-
"learning_rate":
|
| 22 |
-
"loss":
|
| 23 |
"step": 100
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 2.34375,
|
| 27 |
-
"grad_norm":
|
| 28 |
-
"learning_rate":
|
| 29 |
-
"loss":
|
| 30 |
"step": 150
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 3.125,
|
| 34 |
-
"grad_norm":
|
| 35 |
-
"learning_rate":
|
| 36 |
-
"loss": 0.
|
| 37 |
"step": 200
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 3.90625,
|
| 41 |
-
"grad_norm":
|
| 42 |
-
"learning_rate":
|
| 43 |
-
"loss": 0.
|
| 44 |
"step": 250
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 4.6875,
|
| 48 |
-
"grad_norm":
|
| 49 |
-
"learning_rate":
|
| 50 |
-
"loss": 0.
|
| 51 |
"step": 300
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 5.46875,
|
| 55 |
-
"grad_norm":
|
| 56 |
-
"learning_rate":
|
| 57 |
-
"loss": 0.
|
| 58 |
"step": 350
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 6.25,
|
| 62 |
-
"grad_norm":
|
| 63 |
-
"learning_rate":
|
| 64 |
-
"loss": 0.
|
| 65 |
"step": 400
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 7.03125,
|
| 69 |
-
"grad_norm":
|
| 70 |
-
"learning_rate":
|
| 71 |
-
"loss": 0.
|
| 72 |
"step": 450
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 7.8125,
|
| 76 |
-
"grad_norm":
|
| 77 |
-
"learning_rate":
|
| 78 |
-
"loss": 0.
|
| 79 |
"step": 500
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 7.8125,
|
| 83 |
-
"eval_accuracy": 0.
|
| 84 |
-
"eval_loss": 0.
|
| 85 |
-
"eval_runtime":
|
| 86 |
-
"eval_samples_per_second":
|
| 87 |
-
"eval_steps_per_second": 21.
|
| 88 |
"step": 500
|
| 89 |
},
|
| 90 |
{
|
| 91 |
"epoch": 8.59375,
|
| 92 |
-
"grad_norm":
|
| 93 |
-
"learning_rate":
|
| 94 |
-
"loss": 0.
|
| 95 |
"step": 550
|
| 96 |
},
|
| 97 |
{
|
| 98 |
"epoch": 9.375,
|
| 99 |
-
"grad_norm":
|
| 100 |
-
"learning_rate":
|
| 101 |
-
"loss": 0.
|
| 102 |
"step": 600
|
| 103 |
},
|
| 104 |
{
|
| 105 |
"epoch": 10.15625,
|
| 106 |
-
"grad_norm":
|
| 107 |
-
"learning_rate":
|
| 108 |
-
"loss": 0.
|
| 109 |
"step": 650
|
| 110 |
},
|
| 111 |
{
|
| 112 |
"epoch": 10.9375,
|
| 113 |
-
"grad_norm":
|
| 114 |
-
"learning_rate":
|
| 115 |
-
"loss": 0.
|
| 116 |
"step": 700
|
| 117 |
},
|
| 118 |
{
|
| 119 |
"epoch": 11.71875,
|
| 120 |
-
"grad_norm":
|
| 121 |
-
"learning_rate":
|
| 122 |
-
"loss": 0.
|
| 123 |
"step": 750
|
| 124 |
},
|
| 125 |
{
|
| 126 |
"epoch": 12.5,
|
| 127 |
-
"grad_norm":
|
| 128 |
-
"learning_rate":
|
| 129 |
-
"loss": 0.
|
| 130 |
"step": 800
|
| 131 |
},
|
| 132 |
{
|
| 133 |
"epoch": 13.28125,
|
| 134 |
-
"grad_norm":
|
| 135 |
-
"learning_rate":
|
| 136 |
-
"loss": 0.
|
| 137 |
"step": 850
|
| 138 |
},
|
| 139 |
{
|
| 140 |
"epoch": 14.0625,
|
| 141 |
-
"grad_norm":
|
| 142 |
-
"learning_rate":
|
| 143 |
-
"loss": 0.
|
| 144 |
"step": 900
|
| 145 |
},
|
| 146 |
{
|
| 147 |
"epoch": 14.84375,
|
| 148 |
-
"grad_norm":
|
| 149 |
-
"learning_rate":
|
| 150 |
-
"loss": 0.
|
| 151 |
"step": 950
|
| 152 |
},
|
| 153 |
{
|
| 154 |
"epoch": 15.625,
|
| 155 |
-
"grad_norm":
|
| 156 |
-
"learning_rate":
|
| 157 |
-
"loss": 0.
|
| 158 |
"step": 1000
|
| 159 |
},
|
| 160 |
{
|
| 161 |
"epoch": 15.625,
|
| 162 |
-
"eval_accuracy": 0.
|
| 163 |
-
"eval_loss": 0.
|
| 164 |
-
"eval_runtime": 5.
|
| 165 |
-
"eval_samples_per_second":
|
| 166 |
-
"eval_steps_per_second":
|
| 167 |
"step": 1000
|
| 168 |
},
|
| 169 |
{
|
| 170 |
"epoch": 16.40625,
|
| 171 |
-
"grad_norm":
|
| 172 |
-
"learning_rate":
|
| 173 |
-
"loss": 0.
|
| 174 |
"step": 1050
|
| 175 |
},
|
| 176 |
{
|
| 177 |
"epoch": 17.1875,
|
| 178 |
-
"grad_norm": 0.
|
| 179 |
-
"learning_rate":
|
| 180 |
-
"loss": 0.
|
| 181 |
"step": 1100
|
| 182 |
},
|
| 183 |
{
|
| 184 |
"epoch": 17.96875,
|
| 185 |
-
"grad_norm": 0.
|
| 186 |
-
"learning_rate":
|
| 187 |
-
"loss": 0.
|
| 188 |
"step": 1150
|
| 189 |
},
|
| 190 |
{
|
| 191 |
"epoch": 18.75,
|
| 192 |
-
"grad_norm": 0.
|
| 193 |
-
"learning_rate":
|
| 194 |
-
"loss": 0.
|
| 195 |
"step": 1200
|
| 196 |
},
|
| 197 |
{
|
| 198 |
"epoch": 19.53125,
|
| 199 |
-
"grad_norm": 0.
|
| 200 |
-
"learning_rate":
|
| 201 |
-
"loss": 0.
|
| 202 |
"step": 1250
|
| 203 |
},
|
| 204 |
{
|
| 205 |
"epoch": 20.3125,
|
| 206 |
-
"grad_norm": 0.
|
| 207 |
-
"learning_rate":
|
| 208 |
-
"loss": 0.
|
| 209 |
"step": 1300
|
| 210 |
},
|
| 211 |
{
|
| 212 |
"epoch": 21.09375,
|
| 213 |
-
"grad_norm": 0.
|
| 214 |
-
"learning_rate":
|
| 215 |
-
"loss": 0.
|
| 216 |
"step": 1350
|
| 217 |
},
|
| 218 |
{
|
| 219 |
"epoch": 21.875,
|
| 220 |
-
"grad_norm":
|
| 221 |
-
"learning_rate":
|
| 222 |
-
"loss": 0.
|
| 223 |
"step": 1400
|
| 224 |
},
|
| 225 |
{
|
| 226 |
"epoch": 22.65625,
|
| 227 |
-
"grad_norm":
|
| 228 |
-
"learning_rate":
|
| 229 |
-
"loss": 0.
|
| 230 |
"step": 1450
|
| 231 |
},
|
| 232 |
{
|
| 233 |
"epoch": 23.4375,
|
| 234 |
-
"grad_norm": 0.
|
| 235 |
-
"learning_rate":
|
| 236 |
-
"loss": 0.
|
| 237 |
"step": 1500
|
| 238 |
},
|
| 239 |
{
|
| 240 |
"epoch": 23.4375,
|
| 241 |
-
"eval_accuracy": 0.
|
| 242 |
-
"eval_loss": 0.
|
| 243 |
-
"eval_runtime": 5.
|
| 244 |
-
"eval_samples_per_second":
|
| 245 |
-
"eval_steps_per_second":
|
| 246 |
"step": 1500
|
| 247 |
},
|
| 248 |
{
|
| 249 |
"epoch": 24.21875,
|
| 250 |
-
"grad_norm": 0.
|
| 251 |
-
"learning_rate":
|
| 252 |
-
"loss": 0.
|
| 253 |
"step": 1550
|
| 254 |
},
|
| 255 |
{
|
| 256 |
"epoch": 25.0,
|
| 257 |
-
"grad_norm": 0.
|
| 258 |
-
"learning_rate":
|
| 259 |
-
"loss": 0.
|
| 260 |
"step": 1600
|
| 261 |
},
|
| 262 |
{
|
| 263 |
"epoch": 25.78125,
|
| 264 |
-
"grad_norm": 0.
|
| 265 |
-
"learning_rate":
|
| 266 |
-
"loss": 0.
|
| 267 |
"step": 1650
|
| 268 |
},
|
| 269 |
{
|
| 270 |
"epoch": 26.5625,
|
| 271 |
-
"grad_norm": 0.
|
| 272 |
-
"learning_rate":
|
| 273 |
-
"loss": 0.
|
| 274 |
"step": 1700
|
| 275 |
},
|
| 276 |
{
|
| 277 |
"epoch": 27.34375,
|
| 278 |
-
"grad_norm": 0.
|
| 279 |
-
"learning_rate":
|
| 280 |
-
"loss": 0.
|
| 281 |
"step": 1750
|
| 282 |
},
|
| 283 |
{
|
| 284 |
"epoch": 28.125,
|
| 285 |
-
"grad_norm": 0.
|
| 286 |
-
"learning_rate":
|
| 287 |
-
"loss": 0.
|
| 288 |
"step": 1800
|
| 289 |
},
|
| 290 |
{
|
| 291 |
"epoch": 28.90625,
|
| 292 |
-
"grad_norm": 0.
|
| 293 |
-
"learning_rate":
|
| 294 |
-
"loss": 0.
|
| 295 |
"step": 1850
|
| 296 |
},
|
| 297 |
{
|
| 298 |
"epoch": 29.6875,
|
| 299 |
-
"grad_norm": 0.
|
| 300 |
-
"learning_rate":
|
| 301 |
-
"loss": 0.
|
| 302 |
"step": 1900
|
| 303 |
},
|
| 304 |
{
|
| 305 |
"epoch": 30.46875,
|
| 306 |
-
"grad_norm": 0.
|
| 307 |
-
"learning_rate":
|
| 308 |
-
"loss": 0.
|
| 309 |
"step": 1950
|
| 310 |
},
|
| 311 |
{
|
| 312 |
"epoch": 31.25,
|
| 313 |
-
"grad_norm": 0.
|
| 314 |
-
"learning_rate":
|
| 315 |
-
"loss": 0.
|
| 316 |
"step": 2000
|
| 317 |
},
|
| 318 |
{
|
| 319 |
"epoch": 31.25,
|
| 320 |
-
"eval_accuracy": 0.
|
| 321 |
-
"eval_loss": 0.
|
| 322 |
-
"eval_runtime": 5.
|
| 323 |
-
"eval_samples_per_second":
|
| 324 |
-
"eval_steps_per_second":
|
| 325 |
"step": 2000
|
| 326 |
},
|
| 327 |
{
|
| 328 |
"epoch": 32.03125,
|
| 329 |
-
"grad_norm": 0.
|
| 330 |
-
"learning_rate":
|
| 331 |
-
"loss": 0.
|
| 332 |
"step": 2050
|
| 333 |
},
|
| 334 |
{
|
| 335 |
"epoch": 32.8125,
|
| 336 |
-
"grad_norm": 0.
|
| 337 |
-
"learning_rate":
|
| 338 |
-
"loss": 0.
|
| 339 |
"step": 2100
|
| 340 |
},
|
| 341 |
{
|
| 342 |
"epoch": 33.59375,
|
| 343 |
-
"grad_norm": 0.
|
| 344 |
-
"learning_rate":
|
| 345 |
-
"loss": 0.
|
| 346 |
"step": 2150
|
| 347 |
},
|
| 348 |
{
|
| 349 |
"epoch": 34.375,
|
| 350 |
-
"grad_norm": 0.
|
| 351 |
-
"learning_rate":
|
| 352 |
-
"loss": 0.
|
| 353 |
"step": 2200
|
| 354 |
},
|
| 355 |
{
|
| 356 |
"epoch": 35.15625,
|
| 357 |
-
"grad_norm": 0.
|
| 358 |
-
"learning_rate":
|
| 359 |
-
"loss": 0.
|
| 360 |
"step": 2250
|
| 361 |
},
|
| 362 |
{
|
| 363 |
"epoch": 35.9375,
|
| 364 |
-
"grad_norm": 0.
|
| 365 |
-
"learning_rate":
|
| 366 |
-
"loss": 0.
|
| 367 |
"step": 2300
|
| 368 |
},
|
| 369 |
{
|
| 370 |
"epoch": 36.71875,
|
| 371 |
-
"grad_norm": 0.
|
| 372 |
-
"learning_rate":
|
| 373 |
-
"loss": 0.
|
| 374 |
"step": 2350
|
| 375 |
},
|
| 376 |
{
|
| 377 |
"epoch": 37.5,
|
| 378 |
-
"grad_norm": 0.
|
| 379 |
-
"learning_rate":
|
| 380 |
-
"loss": 0.
|
| 381 |
"step": 2400
|
| 382 |
},
|
| 383 |
{
|
| 384 |
"epoch": 38.28125,
|
| 385 |
-
"grad_norm": 0.
|
| 386 |
-
"learning_rate":
|
| 387 |
-
"loss": 0.
|
| 388 |
"step": 2450
|
| 389 |
},
|
| 390 |
{
|
| 391 |
"epoch": 39.0625,
|
| 392 |
-
"grad_norm": 0.
|
| 393 |
-
"learning_rate":
|
| 394 |
-
"loss": 0.
|
| 395 |
"step": 2500
|
| 396 |
},
|
| 397 |
{
|
| 398 |
"epoch": 39.0625,
|
| 399 |
-
"eval_accuracy": 0.
|
| 400 |
-
"eval_loss": 0.
|
| 401 |
-
"eval_runtime":
|
| 402 |
-
"eval_samples_per_second":
|
| 403 |
-
"eval_steps_per_second":
|
| 404 |
"step": 2500
|
| 405 |
},
|
| 406 |
{
|
| 407 |
"epoch": 39.84375,
|
| 408 |
-
"grad_norm": 0.
|
| 409 |
-
"learning_rate":
|
| 410 |
-
"loss": 0.
|
| 411 |
"step": 2550
|
| 412 |
},
|
| 413 |
{
|
| 414 |
"epoch": 40.625,
|
| 415 |
-
"grad_norm": 0.
|
| 416 |
-
"learning_rate":
|
| 417 |
-
"loss": 0.
|
| 418 |
"step": 2600
|
| 419 |
},
|
| 420 |
{
|
| 421 |
"epoch": 41.40625,
|
| 422 |
-
"grad_norm": 0.
|
| 423 |
-
"learning_rate":
|
| 424 |
-
"loss": 0.
|
| 425 |
"step": 2650
|
| 426 |
},
|
| 427 |
{
|
| 428 |
"epoch": 42.1875,
|
| 429 |
-
"grad_norm": 0.
|
| 430 |
-
"learning_rate":
|
| 431 |
-
"loss": 0.
|
| 432 |
"step": 2700
|
| 433 |
},
|
| 434 |
{
|
| 435 |
"epoch": 42.96875,
|
| 436 |
-
"grad_norm": 0.
|
| 437 |
-
"learning_rate":
|
| 438 |
-
"loss": 0.
|
| 439 |
"step": 2750
|
| 440 |
},
|
| 441 |
{
|
| 442 |
"epoch": 43.75,
|
| 443 |
-
"grad_norm": 0.
|
| 444 |
-
"learning_rate":
|
| 445 |
-
"loss": 0.
|
| 446 |
"step": 2800
|
| 447 |
},
|
| 448 |
{
|
| 449 |
"epoch": 44.53125,
|
| 450 |
-
"grad_norm": 0.
|
| 451 |
-
"learning_rate":
|
| 452 |
-
"loss": 0.
|
| 453 |
"step": 2850
|
| 454 |
},
|
| 455 |
{
|
| 456 |
"epoch": 45.3125,
|
| 457 |
-
"grad_norm": 0.
|
| 458 |
-
"learning_rate":
|
| 459 |
-
"loss": 0.
|
| 460 |
"step": 2900
|
| 461 |
},
|
| 462 |
{
|
| 463 |
"epoch": 46.09375,
|
| 464 |
-
"grad_norm": 0.
|
| 465 |
-
"learning_rate":
|
| 466 |
-
"loss": 0.
|
| 467 |
"step": 2950
|
| 468 |
},
|
| 469 |
{
|
| 470 |
"epoch": 46.875,
|
| 471 |
-
"grad_norm": 0.
|
| 472 |
-
"learning_rate":
|
| 473 |
-
"loss": 0.
|
| 474 |
"step": 3000
|
| 475 |
},
|
| 476 |
{
|
| 477 |
"epoch": 46.875,
|
| 478 |
-
"eval_accuracy": 0.
|
| 479 |
-
"eval_loss": 0.
|
| 480 |
-
"eval_runtime": 5.
|
| 481 |
-
"eval_samples_per_second":
|
| 482 |
-
"eval_steps_per_second":
|
| 483 |
"step": 3000
|
| 484 |
},
|
| 485 |
{
|
| 486 |
"epoch": 47.65625,
|
| 487 |
-
"grad_norm": 0.
|
| 488 |
-
"learning_rate":
|
| 489 |
-
"loss": 0.
|
| 490 |
"step": 3050
|
| 491 |
},
|
| 492 |
{
|
| 493 |
"epoch": 48.4375,
|
| 494 |
-
"grad_norm": 0.
|
| 495 |
-
"learning_rate":
|
| 496 |
-
"loss": 0.
|
| 497 |
"step": 3100
|
| 498 |
},
|
| 499 |
{
|
| 500 |
"epoch": 49.21875,
|
| 501 |
-
"grad_norm": 0.
|
| 502 |
-
"learning_rate":
|
| 503 |
-
"loss": 0.
|
| 504 |
"step": 3150
|
| 505 |
},
|
| 506 |
{
|
| 507 |
"epoch": 50.0,
|
| 508 |
-
"grad_norm": 0.
|
| 509 |
-
"learning_rate":
|
| 510 |
-
"loss": 0.
|
| 511 |
"step": 3200
|
| 512 |
},
|
| 513 |
{
|
| 514 |
"epoch": 50.78125,
|
| 515 |
-
"grad_norm": 0.
|
| 516 |
-
"learning_rate":
|
| 517 |
-
"loss": 0.
|
| 518 |
"step": 3250
|
| 519 |
},
|
| 520 |
{
|
| 521 |
"epoch": 51.5625,
|
| 522 |
-
"grad_norm": 0.
|
| 523 |
-
"learning_rate":
|
| 524 |
-
"loss": 0.
|
| 525 |
"step": 3300
|
| 526 |
},
|
| 527 |
{
|
| 528 |
"epoch": 52.34375,
|
| 529 |
-
"grad_norm": 0.
|
| 530 |
-
"learning_rate":
|
| 531 |
-
"loss": 0.
|
| 532 |
"step": 3350
|
| 533 |
},
|
| 534 |
{
|
| 535 |
"epoch": 53.125,
|
| 536 |
-
"grad_norm": 0.
|
| 537 |
-
"learning_rate":
|
| 538 |
-
"loss": 0.
|
| 539 |
"step": 3400
|
| 540 |
},
|
| 541 |
{
|
| 542 |
"epoch": 53.90625,
|
| 543 |
-
"grad_norm": 0.
|
| 544 |
-
"learning_rate":
|
| 545 |
-
"loss": 0.
|
| 546 |
"step": 3450
|
| 547 |
},
|
| 548 |
{
|
| 549 |
"epoch": 54.6875,
|
| 550 |
-
"grad_norm": 0.
|
| 551 |
-
"learning_rate":
|
| 552 |
-
"loss": 0.
|
| 553 |
"step": 3500
|
| 554 |
},
|
| 555 |
{
|
| 556 |
"epoch": 54.6875,
|
| 557 |
-
"eval_accuracy": 0.
|
| 558 |
-
"eval_loss": 0.
|
| 559 |
-
"eval_runtime": 5.
|
| 560 |
-
"eval_samples_per_second":
|
| 561 |
-
"eval_steps_per_second":
|
| 562 |
"step": 3500
|
| 563 |
},
|
| 564 |
{
|
| 565 |
"epoch": 55.46875,
|
| 566 |
-
"grad_norm": 0.
|
| 567 |
-
"learning_rate":
|
| 568 |
-
"loss": 0.
|
| 569 |
"step": 3550
|
| 570 |
},
|
| 571 |
{
|
| 572 |
"epoch": 56.25,
|
| 573 |
-
"grad_norm": 0.
|
| 574 |
-
"learning_rate":
|
| 575 |
-
"loss": 0.
|
| 576 |
"step": 3600
|
| 577 |
},
|
| 578 |
{
|
| 579 |
"epoch": 57.03125,
|
| 580 |
-
"grad_norm": 0.
|
| 581 |
-
"learning_rate":
|
| 582 |
-
"loss": 0.
|
| 583 |
"step": 3650
|
| 584 |
},
|
| 585 |
{
|
| 586 |
"epoch": 57.8125,
|
| 587 |
-
"grad_norm": 0.
|
| 588 |
-
"learning_rate":
|
| 589 |
-
"loss": 0.
|
| 590 |
"step": 3700
|
| 591 |
},
|
| 592 |
{
|
| 593 |
"epoch": 58.59375,
|
| 594 |
-
"grad_norm": 0.
|
| 595 |
-
"learning_rate":
|
| 596 |
-
"loss": 0.
|
| 597 |
"step": 3750
|
| 598 |
},
|
| 599 |
{
|
| 600 |
"epoch": 59.375,
|
| 601 |
-
"grad_norm": 0.
|
| 602 |
-
"learning_rate":
|
| 603 |
-
"loss": 0.
|
| 604 |
"step": 3800
|
| 605 |
},
|
| 606 |
{
|
| 607 |
"epoch": 60.0,
|
| 608 |
"step": 3840,
|
| 609 |
"total_flos": 1.904477274611122e+19,
|
| 610 |
-
"train_loss": 0.
|
| 611 |
-
"train_runtime":
|
| 612 |
-
"train_samples_per_second":
|
| 613 |
-
"train_steps_per_second": 1.
|
| 614 |
}
|
| 615 |
],
|
| 616 |
"logging_steps": 50,
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.1927209496498108,
|
| 3 |
+
"best_model_checkpoint": "./vit-base-brain-tumor-detection3/checkpoint-1500",
|
| 4 |
"epoch": 60.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
"global_step": 3840,
|
|
|
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.78125,
|
| 13 |
+
"grad_norm": 0.014016176573932171,
|
| 14 |
+
"learning_rate": 5.000000000000001e-07,
|
| 15 |
+
"loss": 0.0028,
|
| 16 |
"step": 50
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 1.5625,
|
| 20 |
+
"grad_norm": 0.013060510158538818,
|
| 21 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 22 |
+
"loss": 0.0028,
|
| 23 |
"step": 100
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 2.34375,
|
| 27 |
+
"grad_norm": 0.013337934389710426,
|
| 28 |
+
"learning_rate": 1.5e-06,
|
| 29 |
+
"loss": 0.0028,
|
| 30 |
"step": 150
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 3.125,
|
| 34 |
+
"grad_norm": 0.013658256269991398,
|
| 35 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 36 |
+
"loss": 0.0027,
|
| 37 |
"step": 200
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 3.90625,
|
| 41 |
+
"grad_norm": 0.012454299256205559,
|
| 42 |
+
"learning_rate": 2.5e-06,
|
| 43 |
+
"loss": 0.0026,
|
| 44 |
"step": 250
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 4.6875,
|
| 48 |
+
"grad_norm": 0.013191607780754566,
|
| 49 |
+
"learning_rate": 3e-06,
|
| 50 |
+
"loss": 0.0026,
|
| 51 |
"step": 300
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 5.46875,
|
| 55 |
+
"grad_norm": 0.0118486937135458,
|
| 56 |
+
"learning_rate": 3.5e-06,
|
| 57 |
+
"loss": 0.0025,
|
| 58 |
"step": 350
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 6.25,
|
| 62 |
+
"grad_norm": 0.0345335379242897,
|
| 63 |
+
"learning_rate": 4.000000000000001e-06,
|
| 64 |
+
"loss": 0.0024,
|
| 65 |
"step": 400
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 7.03125,
|
| 69 |
+
"grad_norm": 0.013763554394245148,
|
| 70 |
+
"learning_rate": 4.5e-06,
|
| 71 |
+
"loss": 0.0137,
|
| 72 |
"step": 450
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 7.8125,
|
| 76 |
+
"grad_norm": 0.014735482633113861,
|
| 77 |
+
"learning_rate": 5e-06,
|
| 78 |
+
"loss": 0.0048,
|
| 79 |
"step": 500
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 7.8125,
|
| 83 |
+
"eval_accuracy": 0.947265625,
|
| 84 |
+
"eval_loss": 0.2336536943912506,
|
| 85 |
+
"eval_runtime": 5.8997,
|
| 86 |
+
"eval_samples_per_second": 173.569,
|
| 87 |
+
"eval_steps_per_second": 21.696,
|
| 88 |
"step": 500
|
| 89 |
},
|
| 90 |
{
|
| 91 |
"epoch": 8.59375,
|
| 92 |
+
"grad_norm": 0.013056355528533459,
|
| 93 |
+
"learning_rate": 5.500000000000001e-06,
|
| 94 |
+
"loss": 0.0022,
|
| 95 |
"step": 550
|
| 96 |
},
|
| 97 |
{
|
| 98 |
"epoch": 9.375,
|
| 99 |
+
"grad_norm": 0.009187333285808563,
|
| 100 |
+
"learning_rate": 6e-06,
|
| 101 |
+
"loss": 0.0021,
|
| 102 |
"step": 600
|
| 103 |
},
|
| 104 |
{
|
| 105 |
"epoch": 10.15625,
|
| 106 |
+
"grad_norm": 0.0087556978687644,
|
| 107 |
+
"learning_rate": 6.5000000000000004e-06,
|
| 108 |
+
"loss": 0.0019,
|
| 109 |
"step": 650
|
| 110 |
},
|
| 111 |
{
|
| 112 |
"epoch": 10.9375,
|
| 113 |
+
"grad_norm": 0.008410913869738579,
|
| 114 |
+
"learning_rate": 7e-06,
|
| 115 |
+
"loss": 0.0018,
|
| 116 |
"step": 700
|
| 117 |
},
|
| 118 |
{
|
| 119 |
"epoch": 11.71875,
|
| 120 |
+
"grad_norm": 0.008203917182981968,
|
| 121 |
+
"learning_rate": 7.500000000000001e-06,
|
| 122 |
+
"loss": 0.0017,
|
| 123 |
"step": 750
|
| 124 |
},
|
| 125 |
{
|
| 126 |
"epoch": 12.5,
|
| 127 |
+
"grad_norm": 0.007246215827763081,
|
| 128 |
+
"learning_rate": 8.000000000000001e-06,
|
| 129 |
+
"loss": 0.0016,
|
| 130 |
"step": 800
|
| 131 |
},
|
| 132 |
{
|
| 133 |
"epoch": 13.28125,
|
| 134 |
+
"grad_norm": 0.006727874744683504,
|
| 135 |
+
"learning_rate": 8.5e-06,
|
| 136 |
+
"loss": 0.0015,
|
| 137 |
"step": 850
|
| 138 |
},
|
| 139 |
{
|
| 140 |
"epoch": 14.0625,
|
| 141 |
+
"grad_norm": 0.007697463966906071,
|
| 142 |
+
"learning_rate": 9e-06,
|
| 143 |
+
"loss": 0.0014,
|
| 144 |
"step": 900
|
| 145 |
},
|
| 146 |
{
|
| 147 |
"epoch": 14.84375,
|
| 148 |
+
"grad_norm": 0.005949131678789854,
|
| 149 |
+
"learning_rate": 9.5e-06,
|
| 150 |
+
"loss": 0.0013,
|
| 151 |
"step": 950
|
| 152 |
},
|
| 153 |
{
|
| 154 |
"epoch": 15.625,
|
| 155 |
+
"grad_norm": 0.0054717655293643475,
|
| 156 |
+
"learning_rate": 1e-05,
|
| 157 |
+
"loss": 0.0012,
|
| 158 |
"step": 1000
|
| 159 |
},
|
| 160 |
{
|
| 161 |
"epoch": 15.625,
|
| 162 |
+
"eval_accuracy": 0.953125,
|
| 163 |
+
"eval_loss": 0.19501826167106628,
|
| 164 |
+
"eval_runtime": 5.9147,
|
| 165 |
+
"eval_samples_per_second": 173.128,
|
| 166 |
+
"eval_steps_per_second": 21.641,
|
| 167 |
"step": 1000
|
| 168 |
},
|
| 169 |
{
|
| 170 |
"epoch": 16.40625,
|
| 171 |
+
"grad_norm": 0.005219893530011177,
|
| 172 |
+
"learning_rate": 9.965181058495823e-06,
|
| 173 |
+
"loss": 0.0011,
|
| 174 |
"step": 1050
|
| 175 |
},
|
| 176 |
{
|
| 177 |
"epoch": 17.1875,
|
| 178 |
+
"grad_norm": 0.004757468122988939,
|
| 179 |
+
"learning_rate": 9.930362116991644e-06,
|
| 180 |
+
"loss": 0.0011,
|
| 181 |
"step": 1100
|
| 182 |
},
|
| 183 |
{
|
| 184 |
"epoch": 17.96875,
|
| 185 |
+
"grad_norm": 0.004971610382199287,
|
| 186 |
+
"learning_rate": 9.895543175487466e-06,
|
| 187 |
+
"loss": 0.001,
|
| 188 |
"step": 1150
|
| 189 |
},
|
| 190 |
{
|
| 191 |
"epoch": 18.75,
|
| 192 |
+
"grad_norm": 0.0046828743070364,
|
| 193 |
+
"learning_rate": 9.860724233983288e-06,
|
| 194 |
+
"loss": 0.0009,
|
| 195 |
"step": 1200
|
| 196 |
},
|
| 197 |
{
|
| 198 |
"epoch": 19.53125,
|
| 199 |
+
"grad_norm": 0.004280711989849806,
|
| 200 |
+
"learning_rate": 9.82590529247911e-06,
|
| 201 |
+
"loss": 0.0009,
|
| 202 |
"step": 1250
|
| 203 |
},
|
| 204 |
{
|
| 205 |
"epoch": 20.3125,
|
| 206 |
+
"grad_norm": 0.004425444174557924,
|
| 207 |
+
"learning_rate": 9.79108635097493e-06,
|
| 208 |
+
"loss": 0.0008,
|
| 209 |
"step": 1300
|
| 210 |
},
|
| 211 |
{
|
| 212 |
"epoch": 21.09375,
|
| 213 |
+
"grad_norm": 0.0037732652854174376,
|
| 214 |
+
"learning_rate": 9.756267409470753e-06,
|
| 215 |
+
"loss": 0.0008,
|
| 216 |
"step": 1350
|
| 217 |
},
|
| 218 |
{
|
| 219 |
"epoch": 21.875,
|
| 220 |
+
"grad_norm": 0.0033754699397832155,
|
| 221 |
+
"learning_rate": 9.721448467966575e-06,
|
| 222 |
+
"loss": 0.0007,
|
| 223 |
"step": 1400
|
| 224 |
},
|
| 225 |
{
|
| 226 |
"epoch": 22.65625,
|
| 227 |
+
"grad_norm": 0.003637350630015135,
|
| 228 |
+
"learning_rate": 9.686629526462397e-06,
|
| 229 |
+
"loss": 0.0007,
|
| 230 |
"step": 1450
|
| 231 |
},
|
| 232 |
{
|
| 233 |
"epoch": 23.4375,
|
| 234 |
+
"grad_norm": 0.003412399208173156,
|
| 235 |
+
"learning_rate": 9.651810584958218e-06,
|
| 236 |
+
"loss": 0.0007,
|
| 237 |
"step": 1500
|
| 238 |
},
|
| 239 |
{
|
| 240 |
"epoch": 23.4375,
|
| 241 |
+
"eval_accuracy": 0.9580078125,
|
| 242 |
+
"eval_loss": 0.1927209496498108,
|
| 243 |
+
"eval_runtime": 5.2401,
|
| 244 |
+
"eval_samples_per_second": 195.416,
|
| 245 |
+
"eval_steps_per_second": 24.427,
|
| 246 |
"step": 1500
|
| 247 |
},
|
| 248 |
{
|
| 249 |
"epoch": 24.21875,
|
| 250 |
+
"grad_norm": 0.002839893801137805,
|
| 251 |
+
"learning_rate": 9.61699164345404e-06,
|
| 252 |
+
"loss": 0.0006,
|
| 253 |
"step": 1550
|
| 254 |
},
|
| 255 |
{
|
| 256 |
"epoch": 25.0,
|
| 257 |
+
"grad_norm": 0.0031008291989564896,
|
| 258 |
+
"learning_rate": 9.58217270194986e-06,
|
| 259 |
+
"loss": 0.0006,
|
| 260 |
"step": 1600
|
| 261 |
},
|
| 262 |
{
|
| 263 |
"epoch": 25.78125,
|
| 264 |
+
"grad_norm": 0.002541514113545418,
|
| 265 |
+
"learning_rate": 9.547353760445683e-06,
|
| 266 |
+
"loss": 0.0006,
|
| 267 |
"step": 1650
|
| 268 |
},
|
| 269 |
{
|
| 270 |
"epoch": 26.5625,
|
| 271 |
+
"grad_norm": 0.0025104843080043793,
|
| 272 |
+
"learning_rate": 9.512534818941505e-06,
|
| 273 |
+
"loss": 0.0005,
|
| 274 |
"step": 1700
|
| 275 |
},
|
| 276 |
{
|
| 277 |
"epoch": 27.34375,
|
| 278 |
+
"grad_norm": 0.0023143806029111147,
|
| 279 |
+
"learning_rate": 9.477715877437327e-06,
|
| 280 |
+
"loss": 0.0005,
|
| 281 |
"step": 1750
|
| 282 |
},
|
| 283 |
{
|
| 284 |
"epoch": 28.125,
|
| 285 |
+
"grad_norm": 0.0023780674673616886,
|
| 286 |
+
"learning_rate": 9.442896935933148e-06,
|
| 287 |
+
"loss": 0.0005,
|
| 288 |
"step": 1800
|
| 289 |
},
|
| 290 |
{
|
| 291 |
"epoch": 28.90625,
|
| 292 |
+
"grad_norm": 0.002274406375363469,
|
| 293 |
+
"learning_rate": 9.40807799442897e-06,
|
| 294 |
+
"loss": 0.0005,
|
| 295 |
"step": 1850
|
| 296 |
},
|
| 297 |
{
|
| 298 |
"epoch": 29.6875,
|
| 299 |
+
"grad_norm": 0.002076026052236557,
|
| 300 |
+
"learning_rate": 9.373259052924792e-06,
|
| 301 |
+
"loss": 0.0005,
|
| 302 |
"step": 1900
|
| 303 |
},
|
| 304 |
{
|
| 305 |
"epoch": 30.46875,
|
| 306 |
+
"grad_norm": 0.0024436817038804293,
|
| 307 |
+
"learning_rate": 9.338440111420614e-06,
|
| 308 |
+
"loss": 0.0004,
|
| 309 |
"step": 1950
|
| 310 |
},
|
| 311 |
{
|
| 312 |
"epoch": 31.25,
|
| 313 |
+
"grad_norm": 0.0018446892499923706,
|
| 314 |
+
"learning_rate": 9.303621169916436e-06,
|
| 315 |
+
"loss": 0.0004,
|
| 316 |
"step": 2000
|
| 317 |
},
|
| 318 |
{
|
| 319 |
"epoch": 31.25,
|
| 320 |
+
"eval_accuracy": 0.962890625,
|
| 321 |
+
"eval_loss": 0.1969820261001587,
|
| 322 |
+
"eval_runtime": 5.2387,
|
| 323 |
+
"eval_samples_per_second": 195.469,
|
| 324 |
+
"eval_steps_per_second": 24.434,
|
| 325 |
"step": 2000
|
| 326 |
},
|
| 327 |
{
|
| 328 |
"epoch": 32.03125,
|
| 329 |
+
"grad_norm": 0.0020159403793513775,
|
| 330 |
+
"learning_rate": 9.268802228412257e-06,
|
| 331 |
+
"loss": 0.0004,
|
| 332 |
"step": 2050
|
| 333 |
},
|
| 334 |
{
|
| 335 |
"epoch": 32.8125,
|
| 336 |
+
"grad_norm": 0.0019202978583052754,
|
| 337 |
+
"learning_rate": 9.23398328690808e-06,
|
| 338 |
+
"loss": 0.0004,
|
| 339 |
"step": 2100
|
| 340 |
},
|
| 341 |
{
|
| 342 |
"epoch": 33.59375,
|
| 343 |
+
"grad_norm": 0.0030681404750794172,
|
| 344 |
+
"learning_rate": 9.1991643454039e-06,
|
| 345 |
+
"loss": 0.0004,
|
| 346 |
"step": 2150
|
| 347 |
},
|
| 348 |
{
|
| 349 |
"epoch": 34.375,
|
| 350 |
+
"grad_norm": 0.0016341815935447812,
|
| 351 |
+
"learning_rate": 9.164345403899722e-06,
|
| 352 |
+
"loss": 0.0004,
|
| 353 |
"step": 2200
|
| 354 |
},
|
| 355 |
{
|
| 356 |
"epoch": 35.15625,
|
| 357 |
+
"grad_norm": 0.0016691142227500677,
|
| 358 |
+
"learning_rate": 9.129526462395544e-06,
|
| 359 |
+
"loss": 0.0003,
|
| 360 |
"step": 2250
|
| 361 |
},
|
| 362 |
{
|
| 363 |
"epoch": 35.9375,
|
| 364 |
+
"grad_norm": 0.0017921621911227703,
|
| 365 |
+
"learning_rate": 9.094707520891366e-06,
|
| 366 |
+
"loss": 0.0003,
|
| 367 |
"step": 2300
|
| 368 |
},
|
| 369 |
{
|
| 370 |
"epoch": 36.71875,
|
| 371 |
+
"grad_norm": 0.00160547427367419,
|
| 372 |
+
"learning_rate": 9.059888579387187e-06,
|
| 373 |
+
"loss": 0.0003,
|
| 374 |
"step": 2350
|
| 375 |
},
|
| 376 |
{
|
| 377 |
"epoch": 37.5,
|
| 378 |
+
"grad_norm": 0.0014217059360817075,
|
| 379 |
+
"learning_rate": 9.025069637883009e-06,
|
| 380 |
+
"loss": 0.0003,
|
| 381 |
"step": 2400
|
| 382 |
},
|
| 383 |
{
|
| 384 |
"epoch": 38.28125,
|
| 385 |
+
"grad_norm": 0.001448018359951675,
|
| 386 |
+
"learning_rate": 8.990250696378831e-06,
|
| 387 |
+
"loss": 0.0003,
|
| 388 |
"step": 2450
|
| 389 |
},
|
| 390 |
{
|
| 391 |
"epoch": 39.0625,
|
| 392 |
+
"grad_norm": 0.0017675248673185706,
|
| 393 |
+
"learning_rate": 8.955431754874653e-06,
|
| 394 |
+
"loss": 0.0003,
|
| 395 |
"step": 2500
|
| 396 |
},
|
| 397 |
{
|
| 398 |
"epoch": 39.0625,
|
| 399 |
+
"eval_accuracy": 0.962890625,
|
| 400 |
+
"eval_loss": 0.20403626561164856,
|
| 401 |
+
"eval_runtime": 5.1962,
|
| 402 |
+
"eval_samples_per_second": 197.067,
|
| 403 |
+
"eval_steps_per_second": 24.633,
|
| 404 |
"step": 2500
|
| 405 |
},
|
| 406 |
{
|
| 407 |
"epoch": 39.84375,
|
| 408 |
+
"grad_norm": 0.0017623680178076029,
|
| 409 |
+
"learning_rate": 8.920612813370474e-06,
|
| 410 |
+
"loss": 0.0003,
|
| 411 |
"step": 2550
|
| 412 |
},
|
| 413 |
{
|
| 414 |
"epoch": 40.625,
|
| 415 |
+
"grad_norm": 0.0011810092255473137,
|
| 416 |
+
"learning_rate": 8.885793871866296e-06,
|
| 417 |
+
"loss": 0.0003,
|
| 418 |
"step": 2600
|
| 419 |
},
|
| 420 |
{
|
| 421 |
"epoch": 41.40625,
|
| 422 |
+
"grad_norm": 0.001152553828433156,
|
| 423 |
+
"learning_rate": 8.850974930362117e-06,
|
| 424 |
+
"loss": 0.0003,
|
| 425 |
"step": 2650
|
| 426 |
},
|
| 427 |
{
|
| 428 |
"epoch": 42.1875,
|
| 429 |
+
"grad_norm": 0.0012170104309916496,
|
| 430 |
+
"learning_rate": 8.816155988857939e-06,
|
| 431 |
+
"loss": 0.0003,
|
| 432 |
"step": 2700
|
| 433 |
},
|
| 434 |
{
|
| 435 |
"epoch": 42.96875,
|
| 436 |
+
"grad_norm": 0.0010642099659889936,
|
| 437 |
+
"learning_rate": 8.781337047353761e-06,
|
| 438 |
+
"loss": 0.0002,
|
| 439 |
"step": 2750
|
| 440 |
},
|
| 441 |
{
|
| 442 |
"epoch": 43.75,
|
| 443 |
+
"grad_norm": 0.0010462955106049776,
|
| 444 |
+
"learning_rate": 8.746518105849583e-06,
|
| 445 |
+
"loss": 0.0002,
|
| 446 |
"step": 2800
|
| 447 |
},
|
| 448 |
{
|
| 449 |
"epoch": 44.53125,
|
| 450 |
+
"grad_norm": 0.0010893407743424177,
|
| 451 |
+
"learning_rate": 8.711699164345404e-06,
|
| 452 |
+
"loss": 0.0002,
|
| 453 |
"step": 2850
|
| 454 |
},
|
| 455 |
{
|
| 456 |
"epoch": 45.3125,
|
| 457 |
+
"grad_norm": 0.0010920371860265732,
|
| 458 |
+
"learning_rate": 8.676880222841226e-06,
|
| 459 |
+
"loss": 0.0002,
|
| 460 |
"step": 2900
|
| 461 |
},
|
| 462 |
{
|
| 463 |
"epoch": 46.09375,
|
| 464 |
+
"grad_norm": 0.0010040885536000133,
|
| 465 |
+
"learning_rate": 8.642061281337048e-06,
|
| 466 |
+
"loss": 0.0002,
|
| 467 |
"step": 2950
|
| 468 |
},
|
| 469 |
{
|
| 470 |
"epoch": 46.875,
|
| 471 |
+
"grad_norm": 0.0009422469302080572,
|
| 472 |
+
"learning_rate": 8.60724233983287e-06,
|
| 473 |
+
"loss": 0.0002,
|
| 474 |
"step": 3000
|
| 475 |
},
|
| 476 |
{
|
| 477 |
"epoch": 46.875,
|
| 478 |
+
"eval_accuracy": 0.962890625,
|
| 479 |
+
"eval_loss": 0.21138769388198853,
|
| 480 |
+
"eval_runtime": 5.8076,
|
| 481 |
+
"eval_samples_per_second": 176.32,
|
| 482 |
+
"eval_steps_per_second": 22.04,
|
| 483 |
"step": 3000
|
| 484 |
},
|
| 485 |
{
|
| 486 |
"epoch": 47.65625,
|
| 487 |
+
"grad_norm": 0.0011073002824559808,
|
| 488 |
+
"learning_rate": 8.572423398328693e-06,
|
| 489 |
+
"loss": 0.0002,
|
| 490 |
"step": 3050
|
| 491 |
},
|
| 492 |
{
|
| 493 |
"epoch": 48.4375,
|
| 494 |
+
"grad_norm": 0.000991741195321083,
|
| 495 |
+
"learning_rate": 8.537604456824513e-06,
|
| 496 |
+
"loss": 0.0002,
|
| 497 |
"step": 3100
|
| 498 |
},
|
| 499 |
{
|
| 500 |
"epoch": 49.21875,
|
| 501 |
+
"grad_norm": 0.0008712337585166097,
|
| 502 |
+
"learning_rate": 8.502785515320335e-06,
|
| 503 |
+
"loss": 0.0002,
|
| 504 |
"step": 3150
|
| 505 |
},
|
| 506 |
{
|
| 507 |
"epoch": 50.0,
|
| 508 |
+
"grad_norm": 0.0008826220873743296,
|
| 509 |
+
"learning_rate": 8.467966573816156e-06,
|
| 510 |
+
"loss": 0.0002,
|
| 511 |
"step": 3200
|
| 512 |
},
|
| 513 |
{
|
| 514 |
"epoch": 50.78125,
|
| 515 |
+
"grad_norm": 0.0009179635089822114,
|
| 516 |
+
"learning_rate": 8.433147632311978e-06,
|
| 517 |
+
"loss": 0.0002,
|
| 518 |
"step": 3250
|
| 519 |
},
|
| 520 |
{
|
| 521 |
"epoch": 51.5625,
|
| 522 |
+
"grad_norm": 0.0008320676279254258,
|
| 523 |
+
"learning_rate": 8.3983286908078e-06,
|
| 524 |
+
"loss": 0.0002,
|
| 525 |
"step": 3300
|
| 526 |
},
|
| 527 |
{
|
| 528 |
"epoch": 52.34375,
|
| 529 |
+
"grad_norm": 0.0007437244057655334,
|
| 530 |
+
"learning_rate": 8.363509749303623e-06,
|
| 531 |
+
"loss": 0.0002,
|
| 532 |
"step": 3350
|
| 533 |
},
|
| 534 |
{
|
| 535 |
"epoch": 53.125,
|
| 536 |
+
"grad_norm": 0.0007439731853082776,
|
| 537 |
+
"learning_rate": 8.328690807799443e-06,
|
| 538 |
+
"loss": 0.0002,
|
| 539 |
"step": 3400
|
| 540 |
},
|
| 541 |
{
|
| 542 |
"epoch": 53.90625,
|
| 543 |
+
"grad_norm": 0.0007023093639872968,
|
| 544 |
+
"learning_rate": 8.293871866295265e-06,
|
| 545 |
+
"loss": 0.0002,
|
| 546 |
"step": 3450
|
| 547 |
},
|
| 548 |
{
|
| 549 |
"epoch": 54.6875,
|
| 550 |
+
"grad_norm": 0.0011785700917243958,
|
| 551 |
+
"learning_rate": 8.259052924791087e-06,
|
| 552 |
+
"loss": 0.0002,
|
| 553 |
"step": 3500
|
| 554 |
},
|
| 555 |
{
|
| 556 |
"epoch": 54.6875,
|
| 557 |
+
"eval_accuracy": 0.96484375,
|
| 558 |
+
"eval_loss": 0.217063769698143,
|
| 559 |
+
"eval_runtime": 5.3451,
|
| 560 |
+
"eval_samples_per_second": 191.577,
|
| 561 |
+
"eval_steps_per_second": 23.947,
|
| 562 |
"step": 3500
|
| 563 |
},
|
| 564 |
{
|
| 565 |
"epoch": 55.46875,
|
| 566 |
+
"grad_norm": 0.0007988162687979639,
|
| 567 |
+
"learning_rate": 8.22423398328691e-06,
|
| 568 |
+
"loss": 0.0001,
|
| 569 |
"step": 3550
|
| 570 |
},
|
| 571 |
{
|
| 572 |
"epoch": 56.25,
|
| 573 |
+
"grad_norm": 0.0009737128275446594,
|
| 574 |
+
"learning_rate": 8.18941504178273e-06,
|
| 575 |
+
"loss": 0.0001,
|
| 576 |
"step": 3600
|
| 577 |
},
|
| 578 |
{
|
| 579 |
"epoch": 57.03125,
|
| 580 |
+
"grad_norm": 0.0006344786379486322,
|
| 581 |
+
"learning_rate": 8.154596100278552e-06,
|
| 582 |
+
"loss": 0.0001,
|
| 583 |
"step": 3650
|
| 584 |
},
|
| 585 |
{
|
| 586 |
"epoch": 57.8125,
|
| 587 |
+
"grad_norm": 0.0009238629718311131,
|
| 588 |
+
"learning_rate": 8.119777158774373e-06,
|
| 589 |
+
"loss": 0.0001,
|
| 590 |
"step": 3700
|
| 591 |
},
|
| 592 |
{
|
| 593 |
"epoch": 58.59375,
|
| 594 |
+
"grad_norm": 0.000863746739923954,
|
| 595 |
+
"learning_rate": 8.084958217270195e-06,
|
| 596 |
+
"loss": 0.0001,
|
| 597 |
"step": 3750
|
| 598 |
},
|
| 599 |
{
|
| 600 |
"epoch": 59.375,
|
| 601 |
+
"grad_norm": 0.0005797584308311343,
|
| 602 |
+
"learning_rate": 8.050139275766017e-06,
|
| 603 |
+
"loss": 0.0001,
|
| 604 |
"step": 3800
|
| 605 |
},
|
| 606 |
{
|
| 607 |
"epoch": 60.0,
|
| 608 |
"step": 3840,
|
| 609 |
"total_flos": 1.904477274611122e+19,
|
| 610 |
+
"train_loss": 0.0010260362852325974,
|
| 611 |
+
"train_runtime": 2806.5312,
|
| 612 |
+
"train_samples_per_second": 87.567,
|
| 613 |
+
"train_steps_per_second": 1.368
|
| 614 |
}
|
| 615 |
],
|
| 616 |
"logging_steps": 50,
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5112
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1c6bd5896a6c68059187803c69a823434e3456afd1c4beb716bf45d591574b6c
|
| 3 |
size 5112
|