Commit
·
a20414e
1
Parent(s):
32d83c7
End of training
Browse files- all_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +370 -0
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.98,
|
| 3 |
+
"total_flos": 2.0000469621477704e+19,
|
| 4 |
+
"train_loss": 0.2682455536162499,
|
| 5 |
+
"train_runtime": 3849.7558,
|
| 6 |
+
"train_samples_per_second": 67.342,
|
| 7 |
+
"train_steps_per_second": 0.131
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.98,
|
| 3 |
+
"total_flos": 2.0000469621477704e+19,
|
| 4 |
+
"train_loss": 0.2682455536162499,
|
| 5 |
+
"train_runtime": 3849.7558,
|
| 6 |
+
"train_samples_per_second": 67.342,
|
| 7 |
+
"train_steps_per_second": 0.131
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,370 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9928844151336341,
|
| 3 |
+
"best_model_checkpoint": "vit-base-aihub_model/checkpoint-505",
|
| 4 |
+
"epoch": 4.975369458128079,
|
| 5 |
+
"global_step": 505,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.1,
|
| 12 |
+
"learning_rate": 9.803921568627451e-06,
|
| 13 |
+
"loss": 1.7678,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.2,
|
| 18 |
+
"learning_rate": 1.9607843137254903e-05,
|
| 19 |
+
"loss": 1.6588,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.3,
|
| 24 |
+
"learning_rate": 2.9411764705882354e-05,
|
| 25 |
+
"loss": 1.4279,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.39,
|
| 30 |
+
"learning_rate": 3.9215686274509805e-05,
|
| 31 |
+
"loss": 1.1053,
|
| 32 |
+
"step": 40
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.49,
|
| 36 |
+
"learning_rate": 4.901960784313725e-05,
|
| 37 |
+
"loss": 0.7867,
|
| 38 |
+
"step": 50
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.59,
|
| 42 |
+
"learning_rate": 4.900881057268723e-05,
|
| 43 |
+
"loss": 0.5811,
|
| 44 |
+
"step": 60
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.69,
|
| 48 |
+
"learning_rate": 4.790748898678414e-05,
|
| 49 |
+
"loss": 0.4555,
|
| 50 |
+
"step": 70
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.79,
|
| 54 |
+
"learning_rate": 4.680616740088106e-05,
|
| 55 |
+
"loss": 0.3709,
|
| 56 |
+
"step": 80
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.89,
|
| 60 |
+
"learning_rate": 4.5704845814977974e-05,
|
| 61 |
+
"loss": 0.3156,
|
| 62 |
+
"step": 90
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.99,
|
| 66 |
+
"learning_rate": 4.46035242290749e-05,
|
| 67 |
+
"loss": 0.2905,
|
| 68 |
+
"step": 100
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 1.0,
|
| 72 |
+
"eval_accuracy": 0.9651162790697675,
|
| 73 |
+
"eval_loss": 0.24726606905460358,
|
| 74 |
+
"eval_runtime": 78.0378,
|
| 75 |
+
"eval_samples_per_second": 73.836,
|
| 76 |
+
"eval_steps_per_second": 0.589,
|
| 77 |
+
"step": 101
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 1.08,
|
| 81 |
+
"learning_rate": 4.3502202643171806e-05,
|
| 82 |
+
"loss": 0.268,
|
| 83 |
+
"step": 110
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 1.18,
|
| 87 |
+
"learning_rate": 4.240088105726873e-05,
|
| 88 |
+
"loss": 0.2355,
|
| 89 |
+
"step": 120
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 1.28,
|
| 93 |
+
"learning_rate": 4.129955947136564e-05,
|
| 94 |
+
"loss": 0.2228,
|
| 95 |
+
"step": 130
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 1.38,
|
| 99 |
+
"learning_rate": 4.019823788546256e-05,
|
| 100 |
+
"loss": 0.2134,
|
| 101 |
+
"step": 140
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 1.48,
|
| 105 |
+
"learning_rate": 3.909691629955947e-05,
|
| 106 |
+
"loss": 0.1868,
|
| 107 |
+
"step": 150
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 1.58,
|
| 111 |
+
"learning_rate": 3.799559471365639e-05,
|
| 112 |
+
"loss": 0.1681,
|
| 113 |
+
"step": 160
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 1.67,
|
| 117 |
+
"learning_rate": 3.689427312775331e-05,
|
| 118 |
+
"loss": 0.1641,
|
| 119 |
+
"step": 170
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 1.77,
|
| 123 |
+
"learning_rate": 3.579295154185022e-05,
|
| 124 |
+
"loss": 0.1532,
|
| 125 |
+
"step": 180
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 1.87,
|
| 129 |
+
"learning_rate": 3.469162995594714e-05,
|
| 130 |
+
"loss": 0.1507,
|
| 131 |
+
"step": 190
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 1.97,
|
| 135 |
+
"learning_rate": 3.3590308370044054e-05,
|
| 136 |
+
"loss": 0.1456,
|
| 137 |
+
"step": 200
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 2.0,
|
| 141 |
+
"eval_accuracy": 0.9807358556056924,
|
| 142 |
+
"eval_loss": 0.1328154057264328,
|
| 143 |
+
"eval_runtime": 77.2608,
|
| 144 |
+
"eval_samples_per_second": 74.579,
|
| 145 |
+
"eval_steps_per_second": 0.595,
|
| 146 |
+
"step": 203
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 2.07,
|
| 150 |
+
"learning_rate": 3.2488986784140976e-05,
|
| 151 |
+
"loss": 0.1406,
|
| 152 |
+
"step": 210
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 2.17,
|
| 156 |
+
"learning_rate": 3.1387665198237885e-05,
|
| 157 |
+
"loss": 0.1294,
|
| 158 |
+
"step": 220
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 2.27,
|
| 162 |
+
"learning_rate": 3.0286343612334804e-05,
|
| 163 |
+
"loss": 0.1316,
|
| 164 |
+
"step": 230
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 2.36,
|
| 168 |
+
"learning_rate": 2.918502202643172e-05,
|
| 169 |
+
"loss": 0.1237,
|
| 170 |
+
"step": 240
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 2.46,
|
| 174 |
+
"learning_rate": 2.8083700440528636e-05,
|
| 175 |
+
"loss": 0.1172,
|
| 176 |
+
"step": 250
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 2.56,
|
| 180 |
+
"learning_rate": 2.698237885462555e-05,
|
| 181 |
+
"loss": 0.1196,
|
| 182 |
+
"step": 260
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 2.66,
|
| 186 |
+
"learning_rate": 2.588105726872247e-05,
|
| 187 |
+
"loss": 0.1054,
|
| 188 |
+
"step": 270
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 2.76,
|
| 192 |
+
"learning_rate": 2.4779735682819386e-05,
|
| 193 |
+
"loss": 0.109,
|
| 194 |
+
"step": 280
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 2.86,
|
| 198 |
+
"learning_rate": 2.3678414096916302e-05,
|
| 199 |
+
"loss": 0.0986,
|
| 200 |
+
"step": 290
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 2.96,
|
| 204 |
+
"learning_rate": 2.2577092511013218e-05,
|
| 205 |
+
"loss": 0.1013,
|
| 206 |
+
"step": 300
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 3.0,
|
| 210 |
+
"eval_accuracy": 0.9925373134328358,
|
| 211 |
+
"eval_loss": 0.08586321026086807,
|
| 212 |
+
"eval_runtime": 76.4972,
|
| 213 |
+
"eval_samples_per_second": 75.323,
|
| 214 |
+
"eval_steps_per_second": 0.601,
|
| 215 |
+
"step": 304
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 3.05,
|
| 219 |
+
"learning_rate": 2.1475770925110134e-05,
|
| 220 |
+
"loss": 0.104,
|
| 221 |
+
"step": 310
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 3.15,
|
| 225 |
+
"learning_rate": 2.037444933920705e-05,
|
| 226 |
+
"loss": 0.0937,
|
| 227 |
+
"step": 320
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 3.25,
|
| 231 |
+
"learning_rate": 1.9273127753303965e-05,
|
| 232 |
+
"loss": 0.0963,
|
| 233 |
+
"step": 330
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 3.35,
|
| 237 |
+
"learning_rate": 1.817180616740088e-05,
|
| 238 |
+
"loss": 0.0937,
|
| 239 |
+
"step": 340
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 3.45,
|
| 243 |
+
"learning_rate": 1.7070484581497796e-05,
|
| 244 |
+
"loss": 0.0886,
|
| 245 |
+
"step": 350
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 3.55,
|
| 249 |
+
"learning_rate": 1.5969162995594712e-05,
|
| 250 |
+
"loss": 0.0889,
|
| 251 |
+
"step": 360
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 3.65,
|
| 255 |
+
"learning_rate": 1.486784140969163e-05,
|
| 256 |
+
"loss": 0.0788,
|
| 257 |
+
"step": 370
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 3.74,
|
| 261 |
+
"learning_rate": 1.3766519823788545e-05,
|
| 262 |
+
"loss": 0.0845,
|
| 263 |
+
"step": 380
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 3.84,
|
| 267 |
+
"learning_rate": 1.2665198237885464e-05,
|
| 268 |
+
"loss": 0.084,
|
| 269 |
+
"step": 390
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 3.94,
|
| 273 |
+
"learning_rate": 1.1563876651982378e-05,
|
| 274 |
+
"loss": 0.0754,
|
| 275 |
+
"step": 400
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 4.0,
|
| 279 |
+
"eval_accuracy": 0.990628254078445,
|
| 280 |
+
"eval_loss": 0.0793519839644432,
|
| 281 |
+
"eval_runtime": 74.187,
|
| 282 |
+
"eval_samples_per_second": 77.669,
|
| 283 |
+
"eval_steps_per_second": 0.62,
|
| 284 |
+
"step": 406
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 4.04,
|
| 288 |
+
"learning_rate": 1.0462555066079296e-05,
|
| 289 |
+
"loss": 0.085,
|
| 290 |
+
"step": 410
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 4.14,
|
| 294 |
+
"learning_rate": 9.361233480176213e-06,
|
| 295 |
+
"loss": 0.0793,
|
| 296 |
+
"step": 420
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 4.24,
|
| 300 |
+
"learning_rate": 8.259911894273129e-06,
|
| 301 |
+
"loss": 0.0826,
|
| 302 |
+
"step": 430
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 4.33,
|
| 306 |
+
"learning_rate": 7.1585903083700445e-06,
|
| 307 |
+
"loss": 0.0813,
|
| 308 |
+
"step": 440
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 4.43,
|
| 312 |
+
"learning_rate": 6.05726872246696e-06,
|
| 313 |
+
"loss": 0.0791,
|
| 314 |
+
"step": 450
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 4.53,
|
| 318 |
+
"learning_rate": 4.955947136563877e-06,
|
| 319 |
+
"loss": 0.0744,
|
| 320 |
+
"step": 460
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 4.63,
|
| 324 |
+
"learning_rate": 3.854625550660793e-06,
|
| 325 |
+
"loss": 0.0752,
|
| 326 |
+
"step": 470
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 4.73,
|
| 330 |
+
"learning_rate": 2.7533039647577095e-06,
|
| 331 |
+
"loss": 0.0756,
|
| 332 |
+
"step": 480
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 4.83,
|
| 336 |
+
"learning_rate": 1.6519823788546256e-06,
|
| 337 |
+
"loss": 0.076,
|
| 338 |
+
"step": 490
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 4.93,
|
| 342 |
+
"learning_rate": 5.506607929515419e-07,
|
| 343 |
+
"loss": 0.0721,
|
| 344 |
+
"step": 500
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 4.98,
|
| 348 |
+
"eval_accuracy": 0.9928844151336341,
|
| 349 |
+
"eval_loss": 0.06965512782335281,
|
| 350 |
+
"eval_runtime": 73.5675,
|
| 351 |
+
"eval_samples_per_second": 78.323,
|
| 352 |
+
"eval_steps_per_second": 0.625,
|
| 353 |
+
"step": 505
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 4.98,
|
| 357 |
+
"step": 505,
|
| 358 |
+
"total_flos": 2.0000469621477704e+19,
|
| 359 |
+
"train_loss": 0.2682455536162499,
|
| 360 |
+
"train_runtime": 3849.7558,
|
| 361 |
+
"train_samples_per_second": 67.342,
|
| 362 |
+
"train_steps_per_second": 0.131
|
| 363 |
+
}
|
| 364 |
+
],
|
| 365 |
+
"max_steps": 505,
|
| 366 |
+
"num_train_epochs": 5,
|
| 367 |
+
"total_flos": 2.0000469621477704e+19,
|
| 368 |
+
"trial_name": null,
|
| 369 |
+
"trial_params": null
|
| 370 |
+
}
|