End of training
Browse files
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_accuracy": 0.9753602975360297,
|
| 4 |
+
"eval_loss": 0.07140230387449265,
|
| 5 |
+
"eval_runtime": 36.6123,
|
| 6 |
+
"eval_samples_per_second": 58.751,
|
| 7 |
+
"eval_steps_per_second": 1.857,
|
| 8 |
+
"total_flos": 4.4917456860202107e+18,
|
| 9 |
+
"train_loss": 0.10242866058618028,
|
| 10 |
+
"train_runtime": 2277.1585,
|
| 11 |
+
"train_samples_per_second": 25.492,
|
| 12 |
+
"train_steps_per_second": 0.199
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_accuracy": 0.9753602975360297,
|
| 4 |
+
"eval_loss": 0.07140230387449265,
|
| 5 |
+
"eval_runtime": 36.6123,
|
| 6 |
+
"eval_samples_per_second": 58.751,
|
| 7 |
+
"eval_steps_per_second": 1.857
|
| 8 |
+
}
|
runs/Jul25_10-54-59_628a4864d230/events.out.tfevents.1690284864.628a4864d230.1652.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:05e27ab569f6910e61eb4b0a5a16aba4127584c3391b7fa99be7a3e6b1581175
|
| 3 |
+
size 411
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 4.4917456860202107e+18,
|
| 4 |
+
"train_loss": 0.10242866058618028,
|
| 5 |
+
"train_runtime": 2277.1585,
|
| 6 |
+
"train_samples_per_second": 25.492,
|
| 7 |
+
"train_steps_per_second": 0.199
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,322 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9753602975360297,
|
| 3 |
+
"best_model_checkpoint": "vit-base-patch16-224-blur_vs_clean/checkpoint-453",
|
| 4 |
+
"epoch": 2.9950413223140497,
|
| 5 |
+
"global_step": 453,
|
| 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": 1.0869565217391305e-05,
|
| 13 |
+
"loss": 0.7653,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.13,
|
| 18 |
+
"learning_rate": 2.173913043478261e-05,
|
| 19 |
+
"loss": 0.6297,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.2,
|
| 24 |
+
"learning_rate": 3.260869565217392e-05,
|
| 25 |
+
"loss": 0.4462,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.26,
|
| 30 |
+
"learning_rate": 4.347826086956522e-05,
|
| 31 |
+
"loss": 0.2414,
|
| 32 |
+
"step": 40
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.33,
|
| 36 |
+
"learning_rate": 4.950859950859951e-05,
|
| 37 |
+
"loss": 0.1249,
|
| 38 |
+
"step": 50
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.4,
|
| 42 |
+
"learning_rate": 4.8280098280098285e-05,
|
| 43 |
+
"loss": 0.1293,
|
| 44 |
+
"step": 60
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.46,
|
| 48 |
+
"learning_rate": 4.705159705159705e-05,
|
| 49 |
+
"loss": 0.1083,
|
| 50 |
+
"step": 70
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.53,
|
| 54 |
+
"learning_rate": 4.582309582309582e-05,
|
| 55 |
+
"loss": 0.0917,
|
| 56 |
+
"step": 80
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.6,
|
| 60 |
+
"learning_rate": 4.4594594594594596e-05,
|
| 61 |
+
"loss": 0.1057,
|
| 62 |
+
"step": 90
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.66,
|
| 66 |
+
"learning_rate": 4.336609336609337e-05,
|
| 67 |
+
"loss": 0.0803,
|
| 68 |
+
"step": 100
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.73,
|
| 72 |
+
"learning_rate": 4.213759213759214e-05,
|
| 73 |
+
"loss": 0.0666,
|
| 74 |
+
"step": 110
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.79,
|
| 78 |
+
"learning_rate": 4.0909090909090915e-05,
|
| 79 |
+
"loss": 0.0832,
|
| 80 |
+
"step": 120
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.86,
|
| 84 |
+
"learning_rate": 3.968058968058968e-05,
|
| 85 |
+
"loss": 0.0867,
|
| 86 |
+
"step": 130
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.93,
|
| 90 |
+
"learning_rate": 3.845208845208845e-05,
|
| 91 |
+
"loss": 0.0812,
|
| 92 |
+
"step": 140
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.99,
|
| 96 |
+
"learning_rate": 3.7223587223587226e-05,
|
| 97 |
+
"loss": 0.0539,
|
| 98 |
+
"step": 150
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 1.0,
|
| 102 |
+
"eval_accuracy": 0.9595536959553695,
|
| 103 |
+
"eval_loss": 0.10777629911899567,
|
| 104 |
+
"eval_runtime": 34.5907,
|
| 105 |
+
"eval_samples_per_second": 62.184,
|
| 106 |
+
"eval_steps_per_second": 1.966,
|
| 107 |
+
"step": 151
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 1.06,
|
| 111 |
+
"learning_rate": 3.5995085995085995e-05,
|
| 112 |
+
"loss": 0.0753,
|
| 113 |
+
"step": 160
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 1.12,
|
| 117 |
+
"learning_rate": 3.476658476658477e-05,
|
| 118 |
+
"loss": 0.0619,
|
| 119 |
+
"step": 170
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 1.19,
|
| 123 |
+
"learning_rate": 3.3538083538083545e-05,
|
| 124 |
+
"loss": 0.0605,
|
| 125 |
+
"step": 180
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 1.26,
|
| 129 |
+
"learning_rate": 3.2309582309582306e-05,
|
| 130 |
+
"loss": 0.0473,
|
| 131 |
+
"step": 190
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 1.32,
|
| 135 |
+
"learning_rate": 3.108108108108108e-05,
|
| 136 |
+
"loss": 0.0672,
|
| 137 |
+
"step": 200
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 1.39,
|
| 141 |
+
"learning_rate": 2.9852579852579853e-05,
|
| 142 |
+
"loss": 0.0626,
|
| 143 |
+
"step": 210
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 1.45,
|
| 147 |
+
"learning_rate": 2.8624078624078625e-05,
|
| 148 |
+
"loss": 0.0704,
|
| 149 |
+
"step": 220
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 1.52,
|
| 153 |
+
"learning_rate": 2.73955773955774e-05,
|
| 154 |
+
"loss": 0.0517,
|
| 155 |
+
"step": 230
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 1.59,
|
| 159 |
+
"learning_rate": 2.616707616707617e-05,
|
| 160 |
+
"loss": 0.0494,
|
| 161 |
+
"step": 240
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 1.65,
|
| 165 |
+
"learning_rate": 2.493857493857494e-05,
|
| 166 |
+
"loss": 0.0603,
|
| 167 |
+
"step": 250
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 1.72,
|
| 171 |
+
"learning_rate": 2.371007371007371e-05,
|
| 172 |
+
"loss": 0.0678,
|
| 173 |
+
"step": 260
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 1.79,
|
| 177 |
+
"learning_rate": 2.2481572481572483e-05,
|
| 178 |
+
"loss": 0.0506,
|
| 179 |
+
"step": 270
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 1.85,
|
| 183 |
+
"learning_rate": 2.1253071253071255e-05,
|
| 184 |
+
"loss": 0.0556,
|
| 185 |
+
"step": 280
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 1.92,
|
| 189 |
+
"learning_rate": 2.0024570024570026e-05,
|
| 190 |
+
"loss": 0.0687,
|
| 191 |
+
"step": 290
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 1.98,
|
| 195 |
+
"learning_rate": 1.8796068796068798e-05,
|
| 196 |
+
"loss": 0.0611,
|
| 197 |
+
"step": 300
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 2.0,
|
| 201 |
+
"eval_accuracy": 0.9697814969781497,
|
| 202 |
+
"eval_loss": 0.08457668125629425,
|
| 203 |
+
"eval_runtime": 33.7533,
|
| 204 |
+
"eval_samples_per_second": 63.727,
|
| 205 |
+
"eval_steps_per_second": 2.015,
|
| 206 |
+
"step": 302
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 2.05,
|
| 210 |
+
"learning_rate": 1.756756756756757e-05,
|
| 211 |
+
"loss": 0.0444,
|
| 212 |
+
"step": 310
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 2.12,
|
| 216 |
+
"learning_rate": 1.6339066339066338e-05,
|
| 217 |
+
"loss": 0.0417,
|
| 218 |
+
"step": 320
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 2.18,
|
| 222 |
+
"learning_rate": 1.5110565110565111e-05,
|
| 223 |
+
"loss": 0.0496,
|
| 224 |
+
"step": 330
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 2.25,
|
| 228 |
+
"learning_rate": 1.3882063882063885e-05,
|
| 229 |
+
"loss": 0.045,
|
| 230 |
+
"step": 340
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 2.31,
|
| 234 |
+
"learning_rate": 1.2653562653562653e-05,
|
| 235 |
+
"loss": 0.0496,
|
| 236 |
+
"step": 350
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 2.38,
|
| 240 |
+
"learning_rate": 1.1425061425061426e-05,
|
| 241 |
+
"loss": 0.0481,
|
| 242 |
+
"step": 360
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 2.45,
|
| 246 |
+
"learning_rate": 1.0196560196560196e-05,
|
| 247 |
+
"loss": 0.0329,
|
| 248 |
+
"step": 370
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 2.51,
|
| 252 |
+
"learning_rate": 8.96805896805897e-06,
|
| 253 |
+
"loss": 0.0424,
|
| 254 |
+
"step": 380
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 2.58,
|
| 258 |
+
"learning_rate": 7.73955773955774e-06,
|
| 259 |
+
"loss": 0.0274,
|
| 260 |
+
"step": 390
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 2.64,
|
| 264 |
+
"learning_rate": 6.511056511056511e-06,
|
| 265 |
+
"loss": 0.0394,
|
| 266 |
+
"step": 400
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 2.71,
|
| 270 |
+
"learning_rate": 5.282555282555283e-06,
|
| 271 |
+
"loss": 0.0522,
|
| 272 |
+
"step": 410
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 2.78,
|
| 276 |
+
"learning_rate": 4.0540540540540545e-06,
|
| 277 |
+
"loss": 0.0331,
|
| 278 |
+
"step": 420
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 2.84,
|
| 282 |
+
"learning_rate": 2.8255528255528258e-06,
|
| 283 |
+
"loss": 0.0275,
|
| 284 |
+
"step": 430
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 2.91,
|
| 288 |
+
"learning_rate": 1.5970515970515972e-06,
|
| 289 |
+
"loss": 0.0374,
|
| 290 |
+
"step": 440
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 2.98,
|
| 294 |
+
"learning_rate": 3.6855036855036856e-07,
|
| 295 |
+
"loss": 0.049,
|
| 296 |
+
"step": 450
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 3.0,
|
| 300 |
+
"eval_accuracy": 0.9753602975360297,
|
| 301 |
+
"eval_loss": 0.07140230387449265,
|
| 302 |
+
"eval_runtime": 35.4449,
|
| 303 |
+
"eval_samples_per_second": 60.686,
|
| 304 |
+
"eval_steps_per_second": 1.918,
|
| 305 |
+
"step": 453
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 3.0,
|
| 309 |
+
"step": 453,
|
| 310 |
+
"total_flos": 4.4917456860202107e+18,
|
| 311 |
+
"train_loss": 0.10242866058618028,
|
| 312 |
+
"train_runtime": 2277.1585,
|
| 313 |
+
"train_samples_per_second": 25.492,
|
| 314 |
+
"train_steps_per_second": 0.199
|
| 315 |
+
}
|
| 316 |
+
],
|
| 317 |
+
"max_steps": 453,
|
| 318 |
+
"num_train_epochs": 3,
|
| 319 |
+
"total_flos": 4.4917456860202107e+18,
|
| 320 |
+
"trial_name": null,
|
| 321 |
+
"trial_params": null
|
| 322 |
+
}
|