Commit
•
28995c2
1
Parent(s):
3ba291c
End of training
Browse files- all_results.json +11 -11
- eval_results.json +6 -6
- train_results.json +6 -6
- trainer_state.json +393 -198
all_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 4.
|
| 8 |
-
"total_flos":
|
| 9 |
-
"train_loss": 0.
|
| 10 |
-
"train_runtime":
|
| 11 |
-
"train_samples_per_second":
|
| 12 |
-
"train_steps_per_second": 0.
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 4.99,
|
| 3 |
+
"eval_accuracy": 0.9851851851851852,
|
| 4 |
+
"eval_loss": 0.044658709317445755,
|
| 5 |
+
"eval_runtime": 25.7468,
|
| 6 |
+
"eval_samples_per_second": 157.301,
|
| 7 |
+
"eval_steps_per_second": 4.933,
|
| 8 |
+
"total_flos": 2.347327658721411e+18,
|
| 9 |
+
"train_loss": 0.1689021515602968,
|
| 10 |
+
"train_runtime": 1731.9704,
|
| 11 |
+
"train_samples_per_second": 54.562,
|
| 12 |
+
"train_steps_per_second": 0.424
|
| 13 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 4.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 4.99,
|
| 3 |
+
"eval_accuracy": 0.9851851851851852,
|
| 4 |
+
"eval_loss": 0.044658709317445755,
|
| 5 |
+
"eval_runtime": 25.7468,
|
| 6 |
+
"eval_samples_per_second": 157.301,
|
| 7 |
+
"eval_steps_per_second": 4.933
|
| 8 |
}
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 4.99,
|
| 3 |
+
"total_flos": 2.347327658721411e+18,
|
| 4 |
+
"train_loss": 0.1689021515602968,
|
| 5 |
+
"train_runtime": 1731.9704,
|
| 6 |
+
"train_samples_per_second": 54.562,
|
| 7 |
+
"train_steps_per_second": 0.424
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,313 +1,508 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "swin-tiny-patch4-window7-224-eurosat/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
-
"global_step":
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
-
"epoch": 0.
|
| 12 |
-
"learning_rate":
|
| 13 |
-
"loss": 2.
|
| 14 |
"step": 10
|
| 15 |
},
|
| 16 |
{
|
| 17 |
-
"epoch": 0.
|
| 18 |
-
"learning_rate":
|
| 19 |
-
"loss":
|
| 20 |
"step": 20
|
| 21 |
},
|
| 22 |
{
|
| 23 |
-
"epoch": 0.
|
| 24 |
-
"learning_rate":
|
| 25 |
-
"loss": 1.
|
| 26 |
"step": 30
|
| 27 |
},
|
| 28 |
{
|
| 29 |
-
"epoch": 0.
|
| 30 |
-
"
|
| 31 |
-
"
|
| 32 |
-
"eval_runtime": 4.8236,
|
| 33 |
-
"eval_samples_per_second": 156.731,
|
| 34 |
-
"eval_steps_per_second": 4.976,
|
| 35 |
-
"step": 33
|
| 36 |
-
},
|
| 37 |
-
{
|
| 38 |
-
"epoch": 1.21,
|
| 39 |
-
"learning_rate": 4.882154882154882e-05,
|
| 40 |
-
"loss": 0.5066,
|
| 41 |
"step": 40
|
| 42 |
},
|
| 43 |
{
|
| 44 |
-
"epoch":
|
| 45 |
-
"learning_rate":
|
| 46 |
-
"loss": 0.
|
| 47 |
"step": 50
|
| 48 |
},
|
| 49 |
{
|
| 50 |
-
"epoch":
|
| 51 |
-
"learning_rate": 4.
|
| 52 |
-
"loss": 0.
|
| 53 |
"step": 60
|
| 54 |
},
|
| 55 |
{
|
| 56 |
-
"epoch":
|
| 57 |
-
"
|
| 58 |
-
"
|
| 59 |
-
"eval_runtime": 4.9657,
|
| 60 |
-
"eval_samples_per_second": 152.245,
|
| 61 |
-
"eval_steps_per_second": 4.833,
|
| 62 |
-
"step": 66
|
| 63 |
-
},
|
| 64 |
-
{
|
| 65 |
-
"epoch": 2.12,
|
| 66 |
-
"learning_rate": 4.3771043771043774e-05,
|
| 67 |
-
"loss": 0.1839,
|
| 68 |
"step": 70
|
| 69 |
},
|
| 70 |
{
|
| 71 |
-
"epoch":
|
| 72 |
-
"learning_rate": 4.
|
| 73 |
-
"loss": 0.
|
| 74 |
"step": 80
|
| 75 |
},
|
| 76 |
{
|
| 77 |
-
"epoch":
|
| 78 |
-
"learning_rate": 4.
|
| 79 |
-
"loss": 0.
|
| 80 |
"step": 90
|
| 81 |
},
|
| 82 |
{
|
| 83 |
-
"epoch":
|
| 84 |
-
"
|
| 85 |
-
"
|
| 86 |
-
"eval_runtime": 4.8701,
|
| 87 |
-
"eval_samples_per_second": 155.233,
|
| 88 |
-
"eval_steps_per_second": 4.928,
|
| 89 |
-
"step": 99
|
| 90 |
-
},
|
| 91 |
-
{
|
| 92 |
-
"epoch": 3.03,
|
| 93 |
-
"learning_rate": 3.872053872053872e-05,
|
| 94 |
-
"loss": 0.105,
|
| 95 |
"step": 100
|
| 96 |
},
|
| 97 |
{
|
| 98 |
-
"epoch":
|
| 99 |
-
"learning_rate":
|
| 100 |
-
"loss": 0.
|
| 101 |
"step": 110
|
| 102 |
},
|
| 103 |
{
|
| 104 |
-
"epoch":
|
| 105 |
-
"learning_rate":
|
| 106 |
-
"loss": 0.
|
| 107 |
"step": 120
|
| 108 |
},
|
| 109 |
{
|
| 110 |
-
"epoch":
|
| 111 |
-
"learning_rate":
|
| 112 |
-
"loss": 0.
|
| 113 |
"step": 130
|
| 114 |
},
|
| 115 |
{
|
| 116 |
-
"epoch":
|
| 117 |
-
"
|
| 118 |
-
"
|
| 119 |
-
"
|
| 120 |
-
"eval_samples_per_second": 118.921,
|
| 121 |
-
"eval_steps_per_second": 3.775,
|
| 122 |
-
"step": 132
|
| 123 |
},
|
| 124 |
{
|
| 125 |
-
"epoch":
|
| 126 |
-
"
|
| 127 |
-
"
|
| 128 |
-
"
|
|
|
|
|
|
|
|
|
|
| 129 |
},
|
| 130 |
{
|
| 131 |
-
"epoch":
|
| 132 |
-
"learning_rate":
|
| 133 |
-
"loss": 0.
|
| 134 |
"step": 150
|
| 135 |
},
|
| 136 |
{
|
| 137 |
-
"epoch":
|
| 138 |
-
"learning_rate":
|
| 139 |
-
"loss": 0.
|
| 140 |
"step": 160
|
| 141 |
},
|
| 142 |
{
|
| 143 |
-
"epoch":
|
| 144 |
-
"
|
| 145 |
-
"
|
| 146 |
-
"eval_runtime": 5.5251,
|
| 147 |
-
"eval_samples_per_second": 136.831,
|
| 148 |
-
"eval_steps_per_second": 4.344,
|
| 149 |
-
"step": 165
|
| 150 |
-
},
|
| 151 |
-
{
|
| 152 |
-
"epoch": 5.15,
|
| 153 |
-
"learning_rate": 2.6936026936026937e-05,
|
| 154 |
-
"loss": 0.0559,
|
| 155 |
"step": 170
|
| 156 |
},
|
| 157 |
{
|
| 158 |
-
"epoch":
|
| 159 |
-
"learning_rate":
|
| 160 |
-
"loss": 0.
|
| 161 |
"step": 180
|
| 162 |
},
|
| 163 |
{
|
| 164 |
-
"epoch":
|
| 165 |
-
"learning_rate":
|
| 166 |
-
"loss": 0.
|
| 167 |
"step": 190
|
| 168 |
},
|
| 169 |
{
|
| 170 |
-
"epoch":
|
| 171 |
-
"
|
| 172 |
-
"
|
| 173 |
-
"eval_runtime": 5.3143,
|
| 174 |
-
"eval_samples_per_second": 142.259,
|
| 175 |
-
"eval_steps_per_second": 4.516,
|
| 176 |
-
"step": 198
|
| 177 |
-
},
|
| 178 |
-
{
|
| 179 |
-
"epoch": 6.06,
|
| 180 |
-
"learning_rate": 2.1885521885521887e-05,
|
| 181 |
-
"loss": 0.0514,
|
| 182 |
"step": 200
|
| 183 |
},
|
| 184 |
{
|
| 185 |
-
"epoch":
|
| 186 |
-
"learning_rate":
|
| 187 |
-
"loss": 0.
|
| 188 |
"step": 210
|
| 189 |
},
|
| 190 |
{
|
| 191 |
-
"epoch":
|
| 192 |
-
"learning_rate":
|
| 193 |
-
"loss": 0.
|
| 194 |
"step": 220
|
| 195 |
},
|
| 196 |
{
|
| 197 |
-
"epoch":
|
| 198 |
-
"learning_rate":
|
| 199 |
-
"loss": 0.
|
| 200 |
"step": 230
|
| 201 |
},
|
| 202 |
{
|
| 203 |
-
"epoch":
|
| 204 |
-
"
|
| 205 |
-
"
|
| 206 |
-
"eval_runtime": 5.0836,
|
| 207 |
-
"eval_samples_per_second": 148.714,
|
| 208 |
-
"eval_steps_per_second": 4.721,
|
| 209 |
-
"step": 231
|
| 210 |
-
},
|
| 211 |
-
{
|
| 212 |
-
"epoch": 7.27,
|
| 213 |
-
"learning_rate": 1.5151515151515153e-05,
|
| 214 |
-
"loss": 0.0263,
|
| 215 |
"step": 240
|
| 216 |
},
|
| 217 |
{
|
| 218 |
-
"epoch": 7
|
| 219 |
-
"learning_rate":
|
| 220 |
-
"loss": 0.
|
| 221 |
"step": 250
|
| 222 |
},
|
| 223 |
{
|
| 224 |
-
"epoch":
|
| 225 |
-
"learning_rate":
|
| 226 |
-
"loss": 0.
|
| 227 |
"step": 260
|
| 228 |
},
|
| 229 |
{
|
| 230 |
-
"epoch":
|
| 231 |
-
"
|
| 232 |
-
"
|
| 233 |
-
"eval_runtime": 5.0567,
|
| 234 |
-
"eval_samples_per_second": 149.505,
|
| 235 |
-
"eval_steps_per_second": 4.746,
|
| 236 |
-
"step": 264
|
| 237 |
-
},
|
| 238 |
-
{
|
| 239 |
-
"epoch": 8.18,
|
| 240 |
-
"learning_rate": 1.0101010101010101e-05,
|
| 241 |
-
"loss": 0.0193,
|
| 242 |
"step": 270
|
| 243 |
},
|
| 244 |
{
|
| 245 |
-
"epoch":
|
| 246 |
-
"learning_rate":
|
| 247 |
-
"loss": 0.
|
| 248 |
"step": 280
|
| 249 |
},
|
| 250 |
{
|
| 251 |
-
"epoch":
|
| 252 |
-
"learning_rate":
|
| 253 |
-
"loss": 0.
|
| 254 |
"step": 290
|
| 255 |
},
|
| 256 |
{
|
| 257 |
-
"epoch":
|
| 258 |
-
"eval_accuracy": 0.
|
| 259 |
-
"eval_loss": 0.
|
| 260 |
-
"eval_runtime":
|
| 261 |
-
"eval_samples_per_second": 156.
|
| 262 |
-
"eval_steps_per_second": 4.
|
| 263 |
-
"step":
|
| 264 |
},
|
| 265 |
{
|
| 266 |
-
"epoch":
|
| 267 |
-
"learning_rate":
|
| 268 |
-
"loss": 0.
|
| 269 |
"step": 300
|
| 270 |
},
|
| 271 |
{
|
| 272 |
-
"epoch":
|
| 273 |
-
"learning_rate": 3.
|
| 274 |
-
"loss": 0.
|
| 275 |
"step": 310
|
| 276 |
},
|
| 277 |
{
|
| 278 |
-
"epoch":
|
| 279 |
-
"learning_rate":
|
| 280 |
-
"loss": 0.
|
| 281 |
"step": 320
|
| 282 |
},
|
| 283 |
{
|
| 284 |
-
"epoch":
|
| 285 |
-
"learning_rate":
|
| 286 |
-
"loss": 0.
|
| 287 |
"step": 330
|
| 288 |
},
|
| 289 |
{
|
| 290 |
-
"epoch":
|
| 291 |
-
"
|
| 292 |
-
"
|
| 293 |
-
"
|
| 294 |
-
|
| 295 |
-
|
| 296 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 297 |
},
|
| 298 |
{
|
| 299 |
-
"epoch":
|
| 300 |
-
"step":
|
| 301 |
-
"total_flos":
|
| 302 |
-
"train_loss": 0.
|
| 303 |
-
"train_runtime":
|
| 304 |
-
"train_samples_per_second":
|
| 305 |
-
"train_steps_per_second": 0.
|
| 306 |
}
|
| 307 |
],
|
| 308 |
-
"max_steps":
|
| 309 |
-
"num_train_epochs":
|
| 310 |
-
"total_flos":
|
| 311 |
"trial_name": null,
|
| 312 |
"trial_params": null
|
| 313 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.9851851851851852,
|
| 3 |
+
"best_model_checkpoint": "swin-tiny-patch4-window7-224-eurosat/checkpoint-588",
|
| 4 |
+
"epoch": 4.99492385786802,
|
| 5 |
+
"global_step": 735,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
+
"epoch": 0.07,
|
| 12 |
+
"learning_rate": 6.7567567567567575e-06,
|
| 13 |
+
"loss": 2.2699,
|
| 14 |
"step": 10
|
| 15 |
},
|
| 16 |
{
|
| 17 |
+
"epoch": 0.14,
|
| 18 |
+
"learning_rate": 1.3513513513513515e-05,
|
| 19 |
+
"loss": 2.0753,
|
| 20 |
"step": 20
|
| 21 |
},
|
| 22 |
{
|
| 23 |
+
"epoch": 0.2,
|
| 24 |
+
"learning_rate": 2.0270270270270273e-05,
|
| 25 |
+
"loss": 1.6254,
|
| 26 |
"step": 30
|
| 27 |
},
|
| 28 |
{
|
| 29 |
+
"epoch": 0.27,
|
| 30 |
+
"learning_rate": 2.702702702702703e-05,
|
| 31 |
+
"loss": 0.9696,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 32 |
"step": 40
|
| 33 |
},
|
| 34 |
{
|
| 35 |
+
"epoch": 0.34,
|
| 36 |
+
"learning_rate": 3.3783783783783784e-05,
|
| 37 |
+
"loss": 0.529,
|
| 38 |
"step": 50
|
| 39 |
},
|
| 40 |
{
|
| 41 |
+
"epoch": 0.41,
|
| 42 |
+
"learning_rate": 4.0540540540540545e-05,
|
| 43 |
+
"loss": 0.358,
|
| 44 |
"step": 60
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.47,
|
| 48 |
+
"learning_rate": 4.72972972972973e-05,
|
| 49 |
+
"loss": 0.24,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 50 |
"step": 70
|
| 51 |
},
|
| 52 |
{
|
| 53 |
+
"epoch": 0.54,
|
| 54 |
+
"learning_rate": 4.9546142208774585e-05,
|
| 55 |
+
"loss": 0.2678,
|
| 56 |
"step": 80
|
| 57 |
},
|
| 58 |
{
|
| 59 |
+
"epoch": 0.61,
|
| 60 |
+
"learning_rate": 4.878971255673222e-05,
|
| 61 |
+
"loss": 0.2537,
|
| 62 |
"step": 90
|
| 63 |
},
|
| 64 |
{
|
| 65 |
+
"epoch": 0.68,
|
| 66 |
+
"learning_rate": 4.803328290468986e-05,
|
| 67 |
+
"loss": 0.1921,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 68 |
"step": 100
|
| 69 |
},
|
| 70 |
{
|
| 71 |
+
"epoch": 0.74,
|
| 72 |
+
"learning_rate": 4.7276853252647505e-05,
|
| 73 |
+
"loss": 0.1691,
|
| 74 |
"step": 110
|
| 75 |
},
|
| 76 |
{
|
| 77 |
+
"epoch": 0.81,
|
| 78 |
+
"learning_rate": 4.652042360060514e-05,
|
| 79 |
+
"loss": 0.1716,
|
| 80 |
"step": 120
|
| 81 |
},
|
| 82 |
{
|
| 83 |
+
"epoch": 0.88,
|
| 84 |
+
"learning_rate": 4.576399394856278e-05,
|
| 85 |
+
"loss": 0.1516,
|
| 86 |
"step": 130
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"epoch": 0.95,
|
| 90 |
+
"learning_rate": 4.5007564296520425e-05,
|
| 91 |
+
"loss": 0.1547,
|
| 92 |
+
"step": 140
|
|
|
|
|
|
|
|
|
|
| 93 |
},
|
| 94 |
{
|
| 95 |
+
"epoch": 0.99,
|
| 96 |
+
"eval_accuracy": 0.9711111111111111,
|
| 97 |
+
"eval_loss": 0.09559287130832672,
|
| 98 |
+
"eval_runtime": 38.4532,
|
| 99 |
+
"eval_samples_per_second": 105.323,
|
| 100 |
+
"eval_steps_per_second": 3.303,
|
| 101 |
+
"step": 147
|
| 102 |
},
|
| 103 |
{
|
| 104 |
+
"epoch": 1.02,
|
| 105 |
+
"learning_rate": 4.425113464447806e-05,
|
| 106 |
+
"loss": 0.1434,
|
| 107 |
"step": 150
|
| 108 |
},
|
| 109 |
{
|
| 110 |
+
"epoch": 1.09,
|
| 111 |
+
"learning_rate": 4.34947049924357e-05,
|
| 112 |
+
"loss": 0.0951,
|
| 113 |
"step": 160
|
| 114 |
},
|
| 115 |
{
|
| 116 |
+
"epoch": 1.16,
|
| 117 |
+
"learning_rate": 4.2738275340393345e-05,
|
| 118 |
+
"loss": 0.1042,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 119 |
"step": 170
|
| 120 |
},
|
| 121 |
{
|
| 122 |
+
"epoch": 1.22,
|
| 123 |
+
"learning_rate": 4.198184568835098e-05,
|
| 124 |
+
"loss": 0.0972,
|
| 125 |
"step": 180
|
| 126 |
},
|
| 127 |
{
|
| 128 |
+
"epoch": 1.29,
|
| 129 |
+
"learning_rate": 4.122541603630862e-05,
|
| 130 |
+
"loss": 0.0714,
|
| 131 |
"step": 190
|
| 132 |
},
|
| 133 |
{
|
| 134 |
+
"epoch": 1.36,
|
| 135 |
+
"learning_rate": 4.0468986384266265e-05,
|
| 136 |
+
"loss": 0.1095,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 137 |
"step": 200
|
| 138 |
},
|
| 139 |
{
|
| 140 |
+
"epoch": 1.43,
|
| 141 |
+
"learning_rate": 3.97125567322239e-05,
|
| 142 |
+
"loss": 0.0779,
|
| 143 |
"step": 210
|
| 144 |
},
|
| 145 |
{
|
| 146 |
+
"epoch": 1.49,
|
| 147 |
+
"learning_rate": 3.895612708018154e-05,
|
| 148 |
+
"loss": 0.0869,
|
| 149 |
"step": 220
|
| 150 |
},
|
| 151 |
{
|
| 152 |
+
"epoch": 1.56,
|
| 153 |
+
"learning_rate": 3.8199697428139184e-05,
|
| 154 |
+
"loss": 0.0936,
|
| 155 |
"step": 230
|
| 156 |
},
|
| 157 |
{
|
| 158 |
+
"epoch": 1.63,
|
| 159 |
+
"learning_rate": 3.744326777609682e-05,
|
| 160 |
+
"loss": 0.092,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 161 |
"step": 240
|
| 162 |
},
|
| 163 |
{
|
| 164 |
+
"epoch": 1.7,
|
| 165 |
+
"learning_rate": 3.668683812405446e-05,
|
| 166 |
+
"loss": 0.1006,
|
| 167 |
"step": 250
|
| 168 |
},
|
| 169 |
{
|
| 170 |
+
"epoch": 1.76,
|
| 171 |
+
"learning_rate": 3.5930408472012104e-05,
|
| 172 |
+
"loss": 0.0909,
|
| 173 |
"step": 260
|
| 174 |
},
|
| 175 |
{
|
| 176 |
+
"epoch": 1.83,
|
| 177 |
+
"learning_rate": 3.517397881996974e-05,
|
| 178 |
+
"loss": 0.0705,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 179 |
"step": 270
|
| 180 |
},
|
| 181 |
{
|
| 182 |
+
"epoch": 1.9,
|
| 183 |
+
"learning_rate": 3.441754916792738e-05,
|
| 184 |
+
"loss": 0.0935,
|
| 185 |
"step": 280
|
| 186 |
},
|
| 187 |
{
|
| 188 |
+
"epoch": 1.97,
|
| 189 |
+
"learning_rate": 3.3661119515885024e-05,
|
| 190 |
+
"loss": 0.0707,
|
| 191 |
"step": 290
|
| 192 |
},
|
| 193 |
{
|
| 194 |
+
"epoch": 1.99,
|
| 195 |
+
"eval_accuracy": 0.9733333333333334,
|
| 196 |
+
"eval_loss": 0.07587520033121109,
|
| 197 |
+
"eval_runtime": 25.8049,
|
| 198 |
+
"eval_samples_per_second": 156.947,
|
| 199 |
+
"eval_steps_per_second": 4.922,
|
| 200 |
+
"step": 294
|
| 201 |
},
|
| 202 |
{
|
| 203 |
+
"epoch": 2.04,
|
| 204 |
+
"learning_rate": 3.290468986384266e-05,
|
| 205 |
+
"loss": 0.0665,
|
| 206 |
"step": 300
|
| 207 |
},
|
| 208 |
{
|
| 209 |
+
"epoch": 2.11,
|
| 210 |
+
"learning_rate": 3.21482602118003e-05,
|
| 211 |
+
"loss": 0.0556,
|
| 212 |
"step": 310
|
| 213 |
},
|
| 214 |
{
|
| 215 |
+
"epoch": 2.18,
|
| 216 |
+
"learning_rate": 3.1391830559757944e-05,
|
| 217 |
+
"loss": 0.0517,
|
| 218 |
"step": 320
|
| 219 |
},
|
| 220 |
{
|
| 221 |
+
"epoch": 2.24,
|
| 222 |
+
"learning_rate": 3.063540090771558e-05,
|
| 223 |
+
"loss": 0.0705,
|
| 224 |
"step": 330
|
| 225 |
},
|
| 226 |
{
|
| 227 |
+
"epoch": 2.31,
|
| 228 |
+
"learning_rate": 2.9878971255673223e-05,
|
| 229 |
+
"loss": 0.071,
|
| 230 |
+
"step": 340
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 2.38,
|
| 234 |
+
"learning_rate": 2.9122541603630864e-05,
|
| 235 |
+
"loss": 0.0466,
|
| 236 |
+
"step": 350
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 2.45,
|
| 240 |
+
"learning_rate": 2.83661119515885e-05,
|
| 241 |
+
"loss": 0.0351,
|
| 242 |
+
"step": 360
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 2.51,
|
| 246 |
+
"learning_rate": 2.7609682299546143e-05,
|
| 247 |
+
"loss": 0.0698,
|
| 248 |
+
"step": 370
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 2.58,
|
| 252 |
+
"learning_rate": 2.6853252647503784e-05,
|
| 253 |
+
"loss": 0.0423,
|
| 254 |
+
"step": 380
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 2.65,
|
| 258 |
+
"learning_rate": 2.609682299546142e-05,
|
| 259 |
+
"loss": 0.0475,
|
| 260 |
+
"step": 390
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 2.72,
|
| 264 |
+
"learning_rate": 2.5340393343419063e-05,
|
| 265 |
+
"loss": 0.0491,
|
| 266 |
+
"step": 400
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 2.79,
|
| 270 |
+
"learning_rate": 2.45839636913767e-05,
|
| 271 |
+
"loss": 0.0403,
|
| 272 |
+
"step": 410
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 2.85,
|
| 276 |
+
"learning_rate": 2.382753403933434e-05,
|
| 277 |
+
"loss": 0.0569,
|
| 278 |
+
"step": 420
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 2.92,
|
| 282 |
+
"learning_rate": 2.3071104387291982e-05,
|
| 283 |
+
"loss": 0.0509,
|
| 284 |
+
"step": 430
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 2.99,
|
| 288 |
+
"learning_rate": 2.231467473524962e-05,
|
| 289 |
+
"loss": 0.0537,
|
| 290 |
+
"step": 440
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 2.99,
|
| 294 |
+
"eval_accuracy": 0.9767901234567902,
|
| 295 |
+
"eval_loss": 0.06802858412265778,
|
| 296 |
+
"eval_runtime": 35.3363,
|
| 297 |
+
"eval_samples_per_second": 114.613,
|
| 298 |
+
"eval_steps_per_second": 3.594,
|
| 299 |
+
"step": 441
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 3.06,
|
| 303 |
+
"learning_rate": 2.155824508320726e-05,
|
| 304 |
+
"loss": 0.0502,
|
| 305 |
+
"step": 450
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 3.13,
|
| 309 |
+
"learning_rate": 2.0801815431164902e-05,
|
| 310 |
+
"loss": 0.0452,
|
| 311 |
+
"step": 460
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 3.2,
|
| 315 |
+
"learning_rate": 2.004538577912254e-05,
|
| 316 |
+
"loss": 0.0352,
|
| 317 |
+
"step": 470
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 3.26,
|
| 321 |
+
"learning_rate": 1.928895612708018e-05,
|
| 322 |
+
"loss": 0.0385,
|
| 323 |
+
"step": 480
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 3.33,
|
| 327 |
+
"learning_rate": 1.8532526475037822e-05,
|
| 328 |
+
"loss": 0.0441,
|
| 329 |
+
"step": 490
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 3.4,
|
| 333 |
+
"learning_rate": 1.777609682299546e-05,
|
| 334 |
+
"loss": 0.0394,
|
| 335 |
+
"step": 500
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 3.47,
|
| 339 |
+
"learning_rate": 1.70196671709531e-05,
|
| 340 |
+
"loss": 0.0351,
|
| 341 |
+
"step": 510
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 3.53,
|
| 345 |
+
"learning_rate": 1.6263237518910742e-05,
|
| 346 |
+
"loss": 0.0251,
|
| 347 |
+
"step": 520
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 3.6,
|
| 351 |
+
"learning_rate": 1.550680786686838e-05,
|
| 352 |
+
"loss": 0.044,
|
| 353 |
+
"step": 530
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 3.67,
|
| 357 |
+
"learning_rate": 1.4750378214826023e-05,
|
| 358 |
+
"loss": 0.0246,
|
| 359 |
+
"step": 540
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 3.74,
|
| 363 |
+
"learning_rate": 1.3993948562783662e-05,
|
| 364 |
+
"loss": 0.0252,
|
| 365 |
+
"step": 550
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 3.81,
|
| 369 |
+
"learning_rate": 1.3237518910741301e-05,
|
| 370 |
+
"loss": 0.0305,
|
| 371 |
+
"step": 560
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 3.87,
|
| 375 |
+
"learning_rate": 1.248108925869894e-05,
|
| 376 |
+
"loss": 0.0219,
|
| 377 |
+
"step": 570
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 3.94,
|
| 381 |
+
"learning_rate": 1.172465960665658e-05,
|
| 382 |
+
"loss": 0.0302,
|
| 383 |
+
"step": 580
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 3.99,
|
| 387 |
+
"eval_accuracy": 0.9851851851851852,
|
| 388 |
+
"eval_loss": 0.044658709317445755,
|
| 389 |
+
"eval_runtime": 25.6563,
|
| 390 |
+
"eval_samples_per_second": 157.856,
|
| 391 |
+
"eval_steps_per_second": 4.95,
|
| 392 |
+
"step": 588
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 4.01,
|
| 396 |
+
"learning_rate": 1.0968229954614221e-05,
|
| 397 |
+
"loss": 0.0154,
|
| 398 |
+
"step": 590
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 4.08,
|
| 402 |
+
"learning_rate": 1.021180030257186e-05,
|
| 403 |
+
"loss": 0.0225,
|
| 404 |
+
"step": 600
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 4.15,
|
| 408 |
+
"learning_rate": 9.4553706505295e-06,
|
| 409 |
+
"loss": 0.0181,
|
| 410 |
+
"step": 610
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 4.22,
|
| 414 |
+
"learning_rate": 8.698940998487141e-06,
|
| 415 |
+
"loss": 0.0199,
|
| 416 |
+
"step": 620
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 4.28,
|
| 420 |
+
"learning_rate": 7.94251134644478e-06,
|
| 421 |
+
"loss": 0.0254,
|
| 422 |
+
"step": 630
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 4.35,
|
| 426 |
+
"learning_rate": 7.186081694402421e-06,
|
| 427 |
+
"loss": 0.0244,
|
| 428 |
+
"step": 640
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 4.42,
|
| 432 |
+
"learning_rate": 6.429652042360061e-06,
|
| 433 |
+
"loss": 0.0181,
|
| 434 |
+
"step": 650
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 4.49,
|
| 438 |
+
"learning_rate": 5.6732223903177e-06,
|
| 439 |
+
"loss": 0.0154,
|
| 440 |
+
"step": 660
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 4.55,
|
| 444 |
+
"learning_rate": 4.9167927382753406e-06,
|
| 445 |
+
"loss": 0.0164,
|
| 446 |
+
"step": 670
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 4.62,
|
| 450 |
+
"learning_rate": 4.16036308623298e-06,
|
| 451 |
+
"loss": 0.0111,
|
| 452 |
+
"step": 680
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 4.69,
|
| 456 |
+
"learning_rate": 3.40393343419062e-06,
|
| 457 |
+
"loss": 0.0199,
|
| 458 |
+
"step": 690
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 4.76,
|
| 462 |
+
"learning_rate": 2.64750378214826e-06,
|
| 463 |
+
"loss": 0.0191,
|
| 464 |
+
"step": 700
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 4.83,
|
| 468 |
+
"learning_rate": 1.8910741301059002e-06,
|
| 469 |
+
"loss": 0.0171,
|
| 470 |
+
"step": 710
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 4.89,
|
| 474 |
+
"learning_rate": 1.13464447806354e-06,
|
| 475 |
+
"loss": 0.0164,
|
| 476 |
+
"step": 720
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 4.96,
|
| 480 |
+
"learning_rate": 3.7821482602118004e-07,
|
| 481 |
+
"loss": 0.0225,
|
| 482 |
+
"step": 730
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 4.99,
|
| 486 |
+
"eval_accuracy": 0.9837037037037037,
|
| 487 |
+
"eval_loss": 0.048928551375865936,
|
| 488 |
+
"eval_runtime": 25.2637,
|
| 489 |
+
"eval_samples_per_second": 160.309,
|
| 490 |
+
"eval_steps_per_second": 5.027,
|
| 491 |
+
"step": 735
|
| 492 |
},
|
| 493 |
{
|
| 494 |
+
"epoch": 4.99,
|
| 495 |
+
"step": 735,
|
| 496 |
+
"total_flos": 2.347327658721411e+18,
|
| 497 |
+
"train_loss": 0.1689021515602968,
|
| 498 |
+
"train_runtime": 1731.9704,
|
| 499 |
+
"train_samples_per_second": 54.562,
|
| 500 |
+
"train_steps_per_second": 0.424
|
| 501 |
}
|
| 502 |
],
|
| 503 |
+
"max_steps": 735,
|
| 504 |
+
"num_train_epochs": 5,
|
| 505 |
+
"total_flos": 2.347327658721411e+18,
|
| 506 |
"trial_name": null,
|
| 507 |
"trial_params": null
|
| 508 |
}
|