cheers again
Browse files- README.md +4 -3
- all_results.json +9 -9
- eval_results.json +5 -5
- runs/Jan24_01-31-50_579caf070ae1/events.out.tfevents.1706060139.579caf070ae1.3225.12 +3 -0
- train_results.json +4 -4
- trainer_state.json +60 -285
README.md
CHANGED
|
@@ -2,6 +2,7 @@
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: google/vit-base-patch16-224-in21k
|
| 4 |
tags:
|
|
|
|
| 5 |
- generated_from_trainer
|
| 6 |
metrics:
|
| 7 |
- accuracy
|
|
@@ -15,10 +16,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# vit-base-beans-tutorial
|
| 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
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
-
- Loss: 0.
|
| 21 |
-
- Accuracy: 0.
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: google/vit-base-patch16-224-in21k
|
| 4 |
tags:
|
| 5 |
+
- image-classification
|
| 6 |
- generated_from_trainer
|
| 7 |
metrics:
|
| 8 |
- accuracy
|
|
|
|
| 16 |
|
| 17 |
# vit-base-beans-tutorial
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the beans dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.5143
|
| 22 |
+
- Accuracy: 0.9609
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
all_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 5.0,
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime": 2.
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 8.
|
| 8 |
"total_flos": 4.006371770595533e+17,
|
| 9 |
-
"train_loss":
|
| 10 |
-
"train_runtime":
|
| 11 |
-
"train_samples_per_second":
|
| 12 |
-
"train_steps_per_second":
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 5.0,
|
| 3 |
+
"eval_accuracy": 0.9849624060150376,
|
| 4 |
+
"eval_loss": 0.16576917469501495,
|
| 5 |
+
"eval_runtime": 2.0958,
|
| 6 |
+
"eval_samples_per_second": 63.46,
|
| 7 |
+
"eval_steps_per_second": 8.111,
|
| 8 |
"total_flos": 4.006371770595533e+17,
|
| 9 |
+
"train_loss": 3.2931016008100974e-06,
|
| 10 |
+
"train_runtime": 143.2513,
|
| 11 |
+
"train_samples_per_second": 36.09,
|
| 12 |
+
"train_steps_per_second": 2.269
|
| 13 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 5.0,
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime": 2.
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 8.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 5.0,
|
| 3 |
+
"eval_accuracy": 0.9849624060150376,
|
| 4 |
+
"eval_loss": 0.16576917469501495,
|
| 5 |
+
"eval_runtime": 2.0958,
|
| 6 |
+
"eval_samples_per_second": 63.46,
|
| 7 |
+
"eval_steps_per_second": 8.111
|
| 8 |
}
|
runs/Jan24_01-31-50_579caf070ae1/events.out.tfevents.1706060139.579caf070ae1.3225.12
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3ca9c618977c4937a7160dc15ff5974824a77c21eceac11ba2b07382ad9e54de
|
| 3 |
+
size 734
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 5.0,
|
| 3 |
"total_flos": 4.006371770595533e+17,
|
| 4 |
-
"train_loss":
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 5.0,
|
| 3 |
"total_flos": 4.006371770595533e+17,
|
| 4 |
+
"train_loss": 3.2931016008100974e-06,
|
| 5 |
+
"train_runtime": 143.2513,
|
| 6 |
+
"train_samples_per_second": 36.09,
|
| 7 |
+
"train_steps_per_second": 2.269
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,469 +1,244 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
"best_model_checkpoint": "./vit-base-beans-tutorial/checkpoint-100",
|
| 4 |
"epoch": 5.0,
|
| 5 |
"eval_steps": 100,
|
| 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": 0.08,
|
| 13 |
-
"learning_rate": 0.00019692307692307696,
|
| 14 |
-
"loss": 0.0004,
|
| 15 |
-
"step": 10
|
| 16 |
-
},
|
| 17 |
{
|
| 18 |
"epoch": 0.15,
|
| 19 |
"learning_rate": 0.00019384615384615385,
|
| 20 |
-
"loss": 0.
|
| 21 |
-
"step":
|
| 22 |
-
},
|
| 23 |
-
{
|
| 24 |
-
"epoch": 0.23,
|
| 25 |
-
"learning_rate": 0.0001907692307692308,
|
| 26 |
-
"loss": 0.0002,
|
| 27 |
-
"step": 30
|
| 28 |
},
|
| 29 |
{
|
| 30 |
"epoch": 0.31,
|
| 31 |
"learning_rate": 0.0001876923076923077,
|
| 32 |
-
"loss": 0.
|
| 33 |
-
"step":
|
| 34 |
-
},
|
| 35 |
-
{
|
| 36 |
-
"epoch": 0.38,
|
| 37 |
-
"learning_rate": 0.00018461538461538463,
|
| 38 |
-
"loss": 0.0001,
|
| 39 |
-
"step": 50
|
| 40 |
},
|
| 41 |
{
|
| 42 |
"epoch": 0.46,
|
| 43 |
"learning_rate": 0.00018153846153846155,
|
| 44 |
-
"loss": 0.
|
| 45 |
-
"step":
|
| 46 |
-
},
|
| 47 |
-
{
|
| 48 |
-
"epoch": 0.54,
|
| 49 |
-
"learning_rate": 0.00017846153846153847,
|
| 50 |
-
"loss": 0.0001,
|
| 51 |
-
"step": 70
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.62,
|
| 55 |
"learning_rate": 0.0001753846153846154,
|
| 56 |
"loss": 0.0,
|
| 57 |
-
"step":
|
| 58 |
-
},
|
| 59 |
-
{
|
| 60 |
-
"epoch": 0.69,
|
| 61 |
-
"learning_rate": 0.00017230769230769234,
|
| 62 |
-
"loss": 0.0,
|
| 63 |
-
"step": 90
|
| 64 |
},
|
| 65 |
{
|
| 66 |
"epoch": 0.77,
|
| 67 |
"learning_rate": 0.00016923076923076923,
|
| 68 |
"loss": 0.0,
|
| 69 |
-
"step":
|
| 70 |
-
},
|
| 71 |
-
{
|
| 72 |
-
"epoch": 0.77,
|
| 73 |
-
"eval_accuracy": 0.9924812030075187,
|
| 74 |
-
"eval_loss": 0.07803888618946075,
|
| 75 |
-
"eval_runtime": 2.3471,
|
| 76 |
-
"eval_samples_per_second": 56.666,
|
| 77 |
-
"eval_steps_per_second": 7.243,
|
| 78 |
-
"step": 100
|
| 79 |
-
},
|
| 80 |
-
{
|
| 81 |
-
"epoch": 0.85,
|
| 82 |
-
"learning_rate": 0.00016615384615384617,
|
| 83 |
-
"loss": 0.0,
|
| 84 |
-
"step": 110
|
| 85 |
},
|
| 86 |
{
|
| 87 |
"epoch": 0.92,
|
| 88 |
"learning_rate": 0.0001630769230769231,
|
| 89 |
"loss": 0.0,
|
| 90 |
-
"step":
|
| 91 |
-
},
|
| 92 |
-
{
|
| 93 |
-
"epoch": 1.0,
|
| 94 |
-
"learning_rate": 0.00016,
|
| 95 |
-
"loss": 0.0,
|
| 96 |
-
"step": 130
|
| 97 |
},
|
| 98 |
{
|
| 99 |
"epoch": 1.08,
|
| 100 |
"learning_rate": 0.00015692307692307693,
|
| 101 |
"loss": 0.0,
|
| 102 |
-
"step":
|
| 103 |
-
},
|
| 104 |
-
{
|
| 105 |
-
"epoch": 1.15,
|
| 106 |
-
"learning_rate": 0.00015384615384615385,
|
| 107 |
-
"loss": 0.0,
|
| 108 |
-
"step": 150
|
| 109 |
},
|
| 110 |
{
|
| 111 |
"epoch": 1.23,
|
| 112 |
"learning_rate": 0.00015076923076923077,
|
| 113 |
"loss": 0.0,
|
| 114 |
-
"step":
|
| 115 |
-
},
|
| 116 |
-
{
|
| 117 |
-
"epoch": 1.31,
|
| 118 |
-
"learning_rate": 0.00014769230769230772,
|
| 119 |
-
"loss": 0.0,
|
| 120 |
-
"step": 170
|
| 121 |
},
|
| 122 |
{
|
| 123 |
"epoch": 1.38,
|
| 124 |
"learning_rate": 0.0001446153846153846,
|
| 125 |
"loss": 0.0,
|
| 126 |
-
"step":
|
| 127 |
-
},
|
| 128 |
-
{
|
| 129 |
-
"epoch": 1.46,
|
| 130 |
-
"learning_rate": 0.00014153846153846156,
|
| 131 |
-
"loss": 0.0,
|
| 132 |
-
"step": 190
|
| 133 |
},
|
| 134 |
{
|
| 135 |
"epoch": 1.54,
|
| 136 |
"learning_rate": 0.00013846153846153847,
|
| 137 |
"loss": 0.0,
|
| 138 |
-
"step":
|
| 139 |
},
|
| 140 |
{
|
| 141 |
"epoch": 1.54,
|
| 142 |
-
"eval_accuracy": 0.
|
| 143 |
-
"eval_loss": 0.
|
| 144 |
-
"eval_runtime":
|
| 145 |
-
"eval_samples_per_second":
|
| 146 |
-
"eval_steps_per_second":
|
| 147 |
-
"step":
|
| 148 |
-
},
|
| 149 |
-
{
|
| 150 |
-
"epoch": 1.62,
|
| 151 |
-
"learning_rate": 0.0001353846153846154,
|
| 152 |
-
"loss": 0.0,
|
| 153 |
-
"step": 210
|
| 154 |
},
|
| 155 |
{
|
| 156 |
"epoch": 1.69,
|
| 157 |
"learning_rate": 0.0001323076923076923,
|
| 158 |
"loss": 0.0,
|
| 159 |
-
"step":
|
| 160 |
-
},
|
| 161 |
-
{
|
| 162 |
-
"epoch": 1.77,
|
| 163 |
-
"learning_rate": 0.00012923076923076923,
|
| 164 |
-
"loss": 0.0,
|
| 165 |
-
"step": 230
|
| 166 |
},
|
| 167 |
{
|
| 168 |
"epoch": 1.85,
|
| 169 |
"learning_rate": 0.00012615384615384615,
|
| 170 |
"loss": 0.0,
|
| 171 |
-
"step":
|
| 172 |
-
},
|
| 173 |
-
{
|
| 174 |
-
"epoch": 1.92,
|
| 175 |
-
"learning_rate": 0.0001230769230769231,
|
| 176 |
-
"loss": 0.0,
|
| 177 |
-
"step": 250
|
| 178 |
},
|
| 179 |
{
|
| 180 |
"epoch": 2.0,
|
| 181 |
"learning_rate": 0.00012,
|
| 182 |
"loss": 0.0,
|
| 183 |
-
"step":
|
| 184 |
-
},
|
| 185 |
-
{
|
| 186 |
-
"epoch": 2.08,
|
| 187 |
-
"learning_rate": 0.00011692307692307694,
|
| 188 |
-
"loss": 0.0,
|
| 189 |
-
"step": 270
|
| 190 |
},
|
| 191 |
{
|
| 192 |
"epoch": 2.15,
|
| 193 |
"learning_rate": 0.00011384615384615384,
|
| 194 |
"loss": 0.0,
|
| 195 |
-
"step":
|
| 196 |
-
},
|
| 197 |
-
{
|
| 198 |
-
"epoch": 2.23,
|
| 199 |
-
"learning_rate": 0.00011076923076923077,
|
| 200 |
-
"loss": 0.0,
|
| 201 |
-
"step": 290
|
| 202 |
},
|
| 203 |
{
|
| 204 |
"epoch": 2.31,
|
| 205 |
"learning_rate": 0.0001076923076923077,
|
| 206 |
"loss": 0.0,
|
| 207 |
-
"step":
|
| 208 |
-
},
|
| 209 |
-
{
|
| 210 |
-
"epoch": 2.31,
|
| 211 |
-
"eval_accuracy": 0.9774436090225563,
|
| 212 |
-
"eval_loss": 0.1845720261335373,
|
| 213 |
-
"eval_runtime": 1.7918,
|
| 214 |
-
"eval_samples_per_second": 74.226,
|
| 215 |
-
"eval_steps_per_second": 9.488,
|
| 216 |
-
"step": 300
|
| 217 |
-
},
|
| 218 |
-
{
|
| 219 |
-
"epoch": 2.38,
|
| 220 |
-
"learning_rate": 0.00010461538461538463,
|
| 221 |
-
"loss": 0.0,
|
| 222 |
-
"step": 310
|
| 223 |
},
|
| 224 |
{
|
| 225 |
"epoch": 2.46,
|
| 226 |
"learning_rate": 0.00010153846153846153,
|
| 227 |
"loss": 0.0,
|
| 228 |
-
"step":
|
| 229 |
-
},
|
| 230 |
-
{
|
| 231 |
-
"epoch": 2.54,
|
| 232 |
-
"learning_rate": 9.846153846153848e-05,
|
| 233 |
-
"loss": 0.0,
|
| 234 |
-
"step": 330
|
| 235 |
},
|
| 236 |
{
|
| 237 |
"epoch": 2.62,
|
| 238 |
"learning_rate": 9.53846153846154e-05,
|
| 239 |
"loss": 0.0,
|
| 240 |
-
"step":
|
| 241 |
-
},
|
| 242 |
-
{
|
| 243 |
-
"epoch": 2.69,
|
| 244 |
-
"learning_rate": 9.230769230769232e-05,
|
| 245 |
-
"loss": 0.0,
|
| 246 |
-
"step": 350
|
| 247 |
},
|
| 248 |
{
|
| 249 |
"epoch": 2.77,
|
| 250 |
"learning_rate": 8.923076923076924e-05,
|
| 251 |
"loss": 0.0,
|
| 252 |
-
"step":
|
| 253 |
-
},
|
| 254 |
-
{
|
| 255 |
-
"epoch": 2.85,
|
| 256 |
-
"learning_rate": 8.615384615384617e-05,
|
| 257 |
-
"loss": 0.0,
|
| 258 |
-
"step": 370
|
| 259 |
},
|
| 260 |
{
|
| 261 |
"epoch": 2.92,
|
| 262 |
"learning_rate": 8.307692307692309e-05,
|
| 263 |
"loss": 0.0,
|
| 264 |
-
"step":
|
| 265 |
-
},
|
| 266 |
-
{
|
| 267 |
-
"epoch": 3.0,
|
| 268 |
-
"learning_rate": 8e-05,
|
| 269 |
-
"loss": 0.0,
|
| 270 |
-
"step": 390
|
| 271 |
},
|
| 272 |
{
|
| 273 |
"epoch": 3.08,
|
| 274 |
"learning_rate": 7.692307692307693e-05,
|
| 275 |
"loss": 0.0,
|
| 276 |
-
"step":
|
| 277 |
},
|
| 278 |
{
|
| 279 |
"epoch": 3.08,
|
| 280 |
-
"eval_accuracy": 0.
|
| 281 |
-
"eval_loss": 0.
|
| 282 |
-
"eval_runtime": 1.
|
| 283 |
-
"eval_samples_per_second":
|
| 284 |
-
"eval_steps_per_second": 9.
|
| 285 |
-
"step":
|
| 286 |
-
},
|
| 287 |
-
{
|
| 288 |
-
"epoch": 3.15,
|
| 289 |
-
"learning_rate": 7.384615384615386e-05,
|
| 290 |
-
"loss": 0.0,
|
| 291 |
-
"step": 410
|
| 292 |
},
|
| 293 |
{
|
| 294 |
"epoch": 3.23,
|
| 295 |
"learning_rate": 7.076923076923078e-05,
|
| 296 |
"loss": 0.0,
|
| 297 |
-
"step":
|
| 298 |
-
},
|
| 299 |
-
{
|
| 300 |
-
"epoch": 3.31,
|
| 301 |
-
"learning_rate": 6.76923076923077e-05,
|
| 302 |
-
"loss": 0.0,
|
| 303 |
-
"step": 430
|
| 304 |
},
|
| 305 |
{
|
| 306 |
"epoch": 3.38,
|
| 307 |
"learning_rate": 6.461538461538462e-05,
|
| 308 |
"loss": 0.0,
|
| 309 |
-
"step":
|
| 310 |
-
},
|
| 311 |
-
{
|
| 312 |
-
"epoch": 3.46,
|
| 313 |
-
"learning_rate": 6.153846153846155e-05,
|
| 314 |
-
"loss": 0.0,
|
| 315 |
-
"step": 450
|
| 316 |
},
|
| 317 |
{
|
| 318 |
"epoch": 3.54,
|
| 319 |
"learning_rate": 5.846153846153847e-05,
|
| 320 |
"loss": 0.0,
|
| 321 |
-
"step":
|
| 322 |
-
},
|
| 323 |
-
{
|
| 324 |
-
"epoch": 3.62,
|
| 325 |
-
"learning_rate": 5.538461538461539e-05,
|
| 326 |
-
"loss": 0.0,
|
| 327 |
-
"step": 470
|
| 328 |
},
|
| 329 |
{
|
| 330 |
"epoch": 3.69,
|
| 331 |
"learning_rate": 5.230769230769231e-05,
|
| 332 |
"loss": 0.0,
|
| 333 |
-
"step":
|
| 334 |
-
},
|
| 335 |
-
{
|
| 336 |
-
"epoch": 3.77,
|
| 337 |
-
"learning_rate": 4.923076923076924e-05,
|
| 338 |
-
"loss": 0.0,
|
| 339 |
-
"step": 490
|
| 340 |
},
|
| 341 |
{
|
| 342 |
"epoch": 3.85,
|
| 343 |
"learning_rate": 4.615384615384616e-05,
|
| 344 |
"loss": 0.0,
|
| 345 |
-
"step":
|
| 346 |
-
},
|
| 347 |
-
{
|
| 348 |
-
"epoch": 3.85,
|
| 349 |
-
"eval_accuracy": 0.9849624060150376,
|
| 350 |
-
"eval_loss": 0.13960087299346924,
|
| 351 |
-
"eval_runtime": 1.797,
|
| 352 |
-
"eval_samples_per_second": 74.011,
|
| 353 |
-
"eval_steps_per_second": 9.46,
|
| 354 |
-
"step": 500
|
| 355 |
-
},
|
| 356 |
-
{
|
| 357 |
-
"epoch": 3.92,
|
| 358 |
-
"learning_rate": 4.3076923076923084e-05,
|
| 359 |
-
"loss": 0.0,
|
| 360 |
-
"step": 510
|
| 361 |
},
|
| 362 |
{
|
| 363 |
"epoch": 4.0,
|
| 364 |
"learning_rate": 4e-05,
|
| 365 |
"loss": 0.0,
|
| 366 |
-
"step":
|
| 367 |
-
},
|
| 368 |
-
{
|
| 369 |
-
"epoch": 4.08,
|
| 370 |
-
"learning_rate": 3.692307692307693e-05,
|
| 371 |
-
"loss": 0.0,
|
| 372 |
-
"step": 530
|
| 373 |
},
|
| 374 |
{
|
| 375 |
"epoch": 4.15,
|
| 376 |
"learning_rate": 3.384615384615385e-05,
|
| 377 |
"loss": 0.0,
|
| 378 |
-
"step":
|
| 379 |
-
},
|
| 380 |
-
{
|
| 381 |
-
"epoch": 4.23,
|
| 382 |
-
"learning_rate": 3.0769230769230774e-05,
|
| 383 |
-
"loss": 0.0,
|
| 384 |
-
"step": 550
|
| 385 |
},
|
| 386 |
{
|
| 387 |
"epoch": 4.31,
|
| 388 |
"learning_rate": 2.7692307692307694e-05,
|
| 389 |
"loss": 0.0,
|
| 390 |
-
"step":
|
| 391 |
-
},
|
| 392 |
-
{
|
| 393 |
-
"epoch": 4.38,
|
| 394 |
-
"learning_rate": 2.461538461538462e-05,
|
| 395 |
-
"loss": 0.0,
|
| 396 |
-
"step": 570
|
| 397 |
},
|
| 398 |
{
|
| 399 |
"epoch": 4.46,
|
| 400 |
"learning_rate": 2.1538461538461542e-05,
|
| 401 |
"loss": 0.0,
|
| 402 |
-
"step":
|
| 403 |
-
},
|
| 404 |
-
{
|
| 405 |
-
"epoch": 4.54,
|
| 406 |
-
"learning_rate": 1.8461538461538465e-05,
|
| 407 |
-
"loss": 0.0,
|
| 408 |
-
"step": 590
|
| 409 |
},
|
| 410 |
{
|
| 411 |
"epoch": 4.62,
|
| 412 |
"learning_rate": 1.5384615384615387e-05,
|
| 413 |
"loss": 0.0,
|
| 414 |
-
"step":
|
| 415 |
},
|
| 416 |
{
|
| 417 |
"epoch": 4.62,
|
| 418 |
"eval_accuracy": 0.9849624060150376,
|
| 419 |
-
"eval_loss": 0.
|
| 420 |
-
"eval_runtime": 1.
|
| 421 |
-
"eval_samples_per_second":
|
| 422 |
-
"eval_steps_per_second": 9.
|
| 423 |
-
"step":
|
| 424 |
-
},
|
| 425 |
-
{
|
| 426 |
-
"epoch": 4.69,
|
| 427 |
-
"learning_rate": 1.230769230769231e-05,
|
| 428 |
-
"loss": 0.0,
|
| 429 |
-
"step": 610
|
| 430 |
},
|
| 431 |
{
|
| 432 |
"epoch": 4.77,
|
| 433 |
"learning_rate": 9.230769230769232e-06,
|
| 434 |
"loss": 0.0,
|
| 435 |
-
"step":
|
| 436 |
-
},
|
| 437 |
-
{
|
| 438 |
-
"epoch": 4.85,
|
| 439 |
-
"learning_rate": 6.153846153846155e-06,
|
| 440 |
-
"loss": 0.0,
|
| 441 |
-
"step": 630
|
| 442 |
},
|
| 443 |
{
|
| 444 |
"epoch": 4.92,
|
| 445 |
"learning_rate": 3.0769230769230774e-06,
|
| 446 |
"loss": 0.0,
|
| 447 |
-
"step":
|
| 448 |
-
},
|
| 449 |
-
{
|
| 450 |
-
"epoch": 5.0,
|
| 451 |
-
"learning_rate": 0.0,
|
| 452 |
-
"loss": 0.0,
|
| 453 |
-
"step": 650
|
| 454 |
},
|
| 455 |
{
|
| 456 |
"epoch": 5.0,
|
| 457 |
-
"step":
|
| 458 |
"total_flos": 4.006371770595533e+17,
|
| 459 |
-
"train_loss":
|
| 460 |
-
"train_runtime":
|
| 461 |
-
"train_samples_per_second":
|
| 462 |
-
"train_steps_per_second":
|
| 463 |
}
|
| 464 |
],
|
| 465 |
"logging_steps": 10,
|
| 466 |
-
"max_steps":
|
| 467 |
"num_train_epochs": 5,
|
| 468 |
"save_steps": 100,
|
| 469 |
"total_flos": 4.006371770595533e+17,
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.16576917469501495,
|
| 3 |
"best_model_checkpoint": "./vit-base-beans-tutorial/checkpoint-100",
|
| 4 |
"epoch": 5.0,
|
| 5 |
"eval_steps": 100,
|
| 6 |
+
"global_step": 325,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 11 |
{
|
| 12 |
"epoch": 0.15,
|
| 13 |
"learning_rate": 0.00019384615384615385,
|
| 14 |
+
"loss": 0.0,
|
| 15 |
+
"step": 10
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 16 |
},
|
| 17 |
{
|
| 18 |
"epoch": 0.31,
|
| 19 |
"learning_rate": 0.0001876923076923077,
|
| 20 |
+
"loss": 0.0,
|
| 21 |
+
"step": 20
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 22 |
},
|
| 23 |
{
|
| 24 |
"epoch": 0.46,
|
| 25 |
"learning_rate": 0.00018153846153846155,
|
| 26 |
+
"loss": 0.0,
|
| 27 |
+
"step": 30
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 28 |
},
|
| 29 |
{
|
| 30 |
"epoch": 0.62,
|
| 31 |
"learning_rate": 0.0001753846153846154,
|
| 32 |
"loss": 0.0,
|
| 33 |
+
"step": 40
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 34 |
},
|
| 35 |
{
|
| 36 |
"epoch": 0.77,
|
| 37 |
"learning_rate": 0.00016923076923076923,
|
| 38 |
"loss": 0.0,
|
| 39 |
+
"step": 50
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 40 |
},
|
| 41 |
{
|
| 42 |
"epoch": 0.92,
|
| 43 |
"learning_rate": 0.0001630769230769231,
|
| 44 |
"loss": 0.0,
|
| 45 |
+
"step": 60
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 1.08,
|
| 49 |
"learning_rate": 0.00015692307692307693,
|
| 50 |
"loss": 0.0,
|
| 51 |
+
"step": 70
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 1.23,
|
| 55 |
"learning_rate": 0.00015076923076923077,
|
| 56 |
"loss": 0.0,
|
| 57 |
+
"step": 80
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 58 |
},
|
| 59 |
{
|
| 60 |
"epoch": 1.38,
|
| 61 |
"learning_rate": 0.0001446153846153846,
|
| 62 |
"loss": 0.0,
|
| 63 |
+
"step": 90
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 64 |
},
|
| 65 |
{
|
| 66 |
"epoch": 1.54,
|
| 67 |
"learning_rate": 0.00013846153846153847,
|
| 68 |
"loss": 0.0,
|
| 69 |
+
"step": 100
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 1.54,
|
| 73 |
+
"eval_accuracy": 0.9849624060150376,
|
| 74 |
+
"eval_loss": 0.16576917469501495,
|
| 75 |
+
"eval_runtime": 2.163,
|
| 76 |
+
"eval_samples_per_second": 61.49,
|
| 77 |
+
"eval_steps_per_second": 7.86,
|
| 78 |
+
"step": 100
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 79 |
},
|
| 80 |
{
|
| 81 |
"epoch": 1.69,
|
| 82 |
"learning_rate": 0.0001323076923076923,
|
| 83 |
"loss": 0.0,
|
| 84 |
+
"step": 110
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 85 |
},
|
| 86 |
{
|
| 87 |
"epoch": 1.85,
|
| 88 |
"learning_rate": 0.00012615384615384615,
|
| 89 |
"loss": 0.0,
|
| 90 |
+
"step": 120
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 91 |
},
|
| 92 |
{
|
| 93 |
"epoch": 2.0,
|
| 94 |
"learning_rate": 0.00012,
|
| 95 |
"loss": 0.0,
|
| 96 |
+
"step": 130
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 97 |
},
|
| 98 |
{
|
| 99 |
"epoch": 2.15,
|
| 100 |
"learning_rate": 0.00011384615384615384,
|
| 101 |
"loss": 0.0,
|
| 102 |
+
"step": 140
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 103 |
},
|
| 104 |
{
|
| 105 |
"epoch": 2.31,
|
| 106 |
"learning_rate": 0.0001076923076923077,
|
| 107 |
"loss": 0.0,
|
| 108 |
+
"step": 150
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 109 |
},
|
| 110 |
{
|
| 111 |
"epoch": 2.46,
|
| 112 |
"learning_rate": 0.00010153846153846153,
|
| 113 |
"loss": 0.0,
|
| 114 |
+
"step": 160
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 2.62,
|
| 118 |
"learning_rate": 9.53846153846154e-05,
|
| 119 |
"loss": 0.0,
|
| 120 |
+
"step": 170
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 121 |
},
|
| 122 |
{
|
| 123 |
"epoch": 2.77,
|
| 124 |
"learning_rate": 8.923076923076924e-05,
|
| 125 |
"loss": 0.0,
|
| 126 |
+
"step": 180
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 127 |
},
|
| 128 |
{
|
| 129 |
"epoch": 2.92,
|
| 130 |
"learning_rate": 8.307692307692309e-05,
|
| 131 |
"loss": 0.0,
|
| 132 |
+
"step": 190
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 133 |
},
|
| 134 |
{
|
| 135 |
"epoch": 3.08,
|
| 136 |
"learning_rate": 7.692307692307693e-05,
|
| 137 |
"loss": 0.0,
|
| 138 |
+
"step": 200
|
| 139 |
},
|
| 140 |
{
|
| 141 |
"epoch": 3.08,
|
| 142 |
+
"eval_accuracy": 0.9849624060150376,
|
| 143 |
+
"eval_loss": 0.17416749894618988,
|
| 144 |
+
"eval_runtime": 1.7693,
|
| 145 |
+
"eval_samples_per_second": 75.171,
|
| 146 |
+
"eval_steps_per_second": 9.608,
|
| 147 |
+
"step": 200
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 148 |
},
|
| 149 |
{
|
| 150 |
"epoch": 3.23,
|
| 151 |
"learning_rate": 7.076923076923078e-05,
|
| 152 |
"loss": 0.0,
|
| 153 |
+
"step": 210
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 154 |
},
|
| 155 |
{
|
| 156 |
"epoch": 3.38,
|
| 157 |
"learning_rate": 6.461538461538462e-05,
|
| 158 |
"loss": 0.0,
|
| 159 |
+
"step": 220
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 160 |
},
|
| 161 |
{
|
| 162 |
"epoch": 3.54,
|
| 163 |
"learning_rate": 5.846153846153847e-05,
|
| 164 |
"loss": 0.0,
|
| 165 |
+
"step": 230
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 166 |
},
|
| 167 |
{
|
| 168 |
"epoch": 3.69,
|
| 169 |
"learning_rate": 5.230769230769231e-05,
|
| 170 |
"loss": 0.0,
|
| 171 |
+
"step": 240
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 172 |
},
|
| 173 |
{
|
| 174 |
"epoch": 3.85,
|
| 175 |
"learning_rate": 4.615384615384616e-05,
|
| 176 |
"loss": 0.0,
|
| 177 |
+
"step": 250
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 178 |
},
|
| 179 |
{
|
| 180 |
"epoch": 4.0,
|
| 181 |
"learning_rate": 4e-05,
|
| 182 |
"loss": 0.0,
|
| 183 |
+
"step": 260
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 184 |
},
|
| 185 |
{
|
| 186 |
"epoch": 4.15,
|
| 187 |
"learning_rate": 3.384615384615385e-05,
|
| 188 |
"loss": 0.0,
|
| 189 |
+
"step": 270
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 190 |
},
|
| 191 |
{
|
| 192 |
"epoch": 4.31,
|
| 193 |
"learning_rate": 2.7692307692307694e-05,
|
| 194 |
"loss": 0.0,
|
| 195 |
+
"step": 280
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 196 |
},
|
| 197 |
{
|
| 198 |
"epoch": 4.46,
|
| 199 |
"learning_rate": 2.1538461538461542e-05,
|
| 200 |
"loss": 0.0,
|
| 201 |
+
"step": 290
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 202 |
},
|
| 203 |
{
|
| 204 |
"epoch": 4.62,
|
| 205 |
"learning_rate": 1.5384615384615387e-05,
|
| 206 |
"loss": 0.0,
|
| 207 |
+
"step": 300
|
| 208 |
},
|
| 209 |
{
|
| 210 |
"epoch": 4.62,
|
| 211 |
"eval_accuracy": 0.9849624060150376,
|
| 212 |
+
"eval_loss": 0.17769162356853485,
|
| 213 |
+
"eval_runtime": 1.7753,
|
| 214 |
+
"eval_samples_per_second": 74.918,
|
| 215 |
+
"eval_steps_per_second": 9.576,
|
| 216 |
+
"step": 300
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 217 |
},
|
| 218 |
{
|
| 219 |
"epoch": 4.77,
|
| 220 |
"learning_rate": 9.230769230769232e-06,
|
| 221 |
"loss": 0.0,
|
| 222 |
+
"step": 310
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 223 |
},
|
| 224 |
{
|
| 225 |
"epoch": 4.92,
|
| 226 |
"learning_rate": 3.0769230769230774e-06,
|
| 227 |
"loss": 0.0,
|
| 228 |
+
"step": 320
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 229 |
},
|
| 230 |
{
|
| 231 |
"epoch": 5.0,
|
| 232 |
+
"step": 325,
|
| 233 |
"total_flos": 4.006371770595533e+17,
|
| 234 |
+
"train_loss": 3.2931016008100974e-06,
|
| 235 |
+
"train_runtime": 143.2513,
|
| 236 |
+
"train_samples_per_second": 36.09,
|
| 237 |
+
"train_steps_per_second": 2.269
|
| 238 |
}
|
| 239 |
],
|
| 240 |
"logging_steps": 10,
|
| 241 |
+
"max_steps": 325,
|
| 242 |
"num_train_epochs": 5,
|
| 243 |
"save_steps": 100,
|
| 244 |
"total_flos": 4.006371770595533e+17,
|