Commit
·
3d4cd0b
1
Parent(s):
9ba6712
End of training
Browse files- all_results.json +13 -0
- eval_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +313 -0
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.98,
|
| 3 |
+
"eval_accuracy": 0.8218390804597702,
|
| 4 |
+
"eval_loss": 0.5200982093811035,
|
| 5 |
+
"eval_runtime": 154.7969,
|
| 6 |
+
"eval_samples_per_second": 3.372,
|
| 7 |
+
"eval_steps_per_second": 0.426,
|
| 8 |
+
"total_flos": 3.483891679221596e+17,
|
| 9 |
+
"train_loss": 0.9891002352379229,
|
| 10 |
+
"train_runtime": 12153.7671,
|
| 11 |
+
"train_samples_per_second": 1.159,
|
| 12 |
+
"train_steps_per_second": 0.036
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.98,
|
| 3 |
+
"eval_accuracy": 0.8218390804597702,
|
| 4 |
+
"eval_loss": 0.5200982093811035,
|
| 5 |
+
"eval_runtime": 154.7969,
|
| 6 |
+
"eval_samples_per_second": 3.372,
|
| 7 |
+
"eval_steps_per_second": 0.426
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.98,
|
| 3 |
+
"total_flos": 3.483891679221596e+17,
|
| 4 |
+
"train_loss": 0.9891002352379229,
|
| 5 |
+
"train_runtime": 12153.7671,
|
| 6 |
+
"train_samples_per_second": 1.159,
|
| 7 |
+
"train_steps_per_second": 0.036
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,313 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8218390804597702,
|
| 3 |
+
"best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-vit/checkpoint-438",
|
| 4 |
+
"epoch": 2.9846678023850086,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 438,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.07,
|
| 13 |
+
"learning_rate": 1.1363636363636365e-05,
|
| 14 |
+
"loss": 2.5718,
|
| 15 |
+
"step": 10
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.14,
|
| 19 |
+
"learning_rate": 2.272727272727273e-05,
|
| 20 |
+
"loss": 2.4304,
|
| 21 |
+
"step": 20
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.2,
|
| 25 |
+
"learning_rate": 3.409090909090909e-05,
|
| 26 |
+
"loss": 2.2567,
|
| 27 |
+
"step": 30
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.27,
|
| 31 |
+
"learning_rate": 4.545454545454546e-05,
|
| 32 |
+
"loss": 1.9423,
|
| 33 |
+
"step": 40
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.34,
|
| 37 |
+
"learning_rate": 4.9238578680203045e-05,
|
| 38 |
+
"loss": 1.5811,
|
| 39 |
+
"step": 50
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.41,
|
| 43 |
+
"learning_rate": 4.7969543147208126e-05,
|
| 44 |
+
"loss": 1.2969,
|
| 45 |
+
"step": 60
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.48,
|
| 49 |
+
"learning_rate": 4.67005076142132e-05,
|
| 50 |
+
"loss": 1.3403,
|
| 51 |
+
"step": 70
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.55,
|
| 55 |
+
"learning_rate": 4.543147208121827e-05,
|
| 56 |
+
"loss": 0.9912,
|
| 57 |
+
"step": 80
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.61,
|
| 61 |
+
"learning_rate": 4.416243654822335e-05,
|
| 62 |
+
"loss": 1.3058,
|
| 63 |
+
"step": 90
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.68,
|
| 67 |
+
"learning_rate": 4.289340101522843e-05,
|
| 68 |
+
"loss": 1.0968,
|
| 69 |
+
"step": 100
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.75,
|
| 73 |
+
"learning_rate": 4.162436548223351e-05,
|
| 74 |
+
"loss": 1.2238,
|
| 75 |
+
"step": 110
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.82,
|
| 79 |
+
"learning_rate": 4.035532994923858e-05,
|
| 80 |
+
"loss": 1.0739,
|
| 81 |
+
"step": 120
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.89,
|
| 85 |
+
"learning_rate": 3.9086294416243655e-05,
|
| 86 |
+
"loss": 1.0948,
|
| 87 |
+
"step": 130
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.95,
|
| 91 |
+
"learning_rate": 3.7817258883248735e-05,
|
| 92 |
+
"loss": 0.967,
|
| 93 |
+
"step": 140
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.99,
|
| 97 |
+
"eval_accuracy": 0.735632183908046,
|
| 98 |
+
"eval_loss": 0.752730667591095,
|
| 99 |
+
"eval_runtime": 221.3941,
|
| 100 |
+
"eval_samples_per_second": 2.358,
|
| 101 |
+
"eval_steps_per_second": 0.298,
|
| 102 |
+
"step": 146
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.02,
|
| 106 |
+
"learning_rate": 3.654822335025381e-05,
|
| 107 |
+
"loss": 0.9675,
|
| 108 |
+
"step": 150
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 1.09,
|
| 112 |
+
"learning_rate": 3.527918781725888e-05,
|
| 113 |
+
"loss": 0.9117,
|
| 114 |
+
"step": 160
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 1.16,
|
| 118 |
+
"learning_rate": 3.401015228426396e-05,
|
| 119 |
+
"loss": 0.8393,
|
| 120 |
+
"step": 170
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 1.23,
|
| 124 |
+
"learning_rate": 3.2741116751269036e-05,
|
| 125 |
+
"loss": 0.7949,
|
| 126 |
+
"step": 180
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 1.29,
|
| 130 |
+
"learning_rate": 3.147208121827411e-05,
|
| 131 |
+
"loss": 0.8543,
|
| 132 |
+
"step": 190
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 1.36,
|
| 136 |
+
"learning_rate": 3.020304568527919e-05,
|
| 137 |
+
"loss": 0.8966,
|
| 138 |
+
"step": 200
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 1.43,
|
| 142 |
+
"learning_rate": 2.8934010152284264e-05,
|
| 143 |
+
"loss": 0.8289,
|
| 144 |
+
"step": 210
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 1.5,
|
| 148 |
+
"learning_rate": 2.766497461928934e-05,
|
| 149 |
+
"loss": 0.8569,
|
| 150 |
+
"step": 220
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 1.57,
|
| 154 |
+
"learning_rate": 2.6395939086294418e-05,
|
| 155 |
+
"loss": 0.8244,
|
| 156 |
+
"step": 230
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 1.64,
|
| 160 |
+
"learning_rate": 2.5126903553299492e-05,
|
| 161 |
+
"loss": 0.8009,
|
| 162 |
+
"step": 240
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 1.7,
|
| 166 |
+
"learning_rate": 2.385786802030457e-05,
|
| 167 |
+
"loss": 0.765,
|
| 168 |
+
"step": 250
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 1.77,
|
| 172 |
+
"learning_rate": 2.2588832487309646e-05,
|
| 173 |
+
"loss": 0.7593,
|
| 174 |
+
"step": 260
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 1.84,
|
| 178 |
+
"learning_rate": 2.1319796954314723e-05,
|
| 179 |
+
"loss": 0.7606,
|
| 180 |
+
"step": 270
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 1.91,
|
| 184 |
+
"learning_rate": 2.0050761421319797e-05,
|
| 185 |
+
"loss": 0.8093,
|
| 186 |
+
"step": 280
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 1.98,
|
| 190 |
+
"learning_rate": 1.8781725888324874e-05,
|
| 191 |
+
"loss": 0.7448,
|
| 192 |
+
"step": 290
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 2.0,
|
| 196 |
+
"eval_accuracy": 0.7988505747126436,
|
| 197 |
+
"eval_loss": 0.5910783410072327,
|
| 198 |
+
"eval_runtime": 144.8329,
|
| 199 |
+
"eval_samples_per_second": 3.604,
|
| 200 |
+
"eval_steps_per_second": 0.456,
|
| 201 |
+
"step": 293
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 2.04,
|
| 205 |
+
"learning_rate": 1.751269035532995e-05,
|
| 206 |
+
"loss": 0.7165,
|
| 207 |
+
"step": 300
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 2.11,
|
| 211 |
+
"learning_rate": 1.6243654822335024e-05,
|
| 212 |
+
"loss": 0.7588,
|
| 213 |
+
"step": 310
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 2.18,
|
| 217 |
+
"learning_rate": 1.4974619289340103e-05,
|
| 218 |
+
"loss": 0.7021,
|
| 219 |
+
"step": 320
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 2.25,
|
| 223 |
+
"learning_rate": 1.3705583756345178e-05,
|
| 224 |
+
"loss": 0.6398,
|
| 225 |
+
"step": 330
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 2.32,
|
| 229 |
+
"learning_rate": 1.2436548223350254e-05,
|
| 230 |
+
"loss": 0.645,
|
| 231 |
+
"step": 340
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 2.39,
|
| 235 |
+
"learning_rate": 1.116751269035533e-05,
|
| 236 |
+
"loss": 0.7154,
|
| 237 |
+
"step": 350
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 2.45,
|
| 241 |
+
"learning_rate": 9.898477157360408e-06,
|
| 242 |
+
"loss": 0.5742,
|
| 243 |
+
"step": 360
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 2.52,
|
| 247 |
+
"learning_rate": 8.629441624365483e-06,
|
| 248 |
+
"loss": 0.6697,
|
| 249 |
+
"step": 370
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 2.59,
|
| 253 |
+
"learning_rate": 7.360406091370558e-06,
|
| 254 |
+
"loss": 0.5898,
|
| 255 |
+
"step": 380
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 2.66,
|
| 259 |
+
"learning_rate": 6.091370558375635e-06,
|
| 260 |
+
"loss": 0.716,
|
| 261 |
+
"step": 390
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 2.73,
|
| 265 |
+
"learning_rate": 4.822335025380711e-06,
|
| 266 |
+
"loss": 0.7302,
|
| 267 |
+
"step": 400
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 2.79,
|
| 271 |
+
"learning_rate": 3.5532994923857873e-06,
|
| 272 |
+
"loss": 0.5301,
|
| 273 |
+
"step": 410
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 2.86,
|
| 277 |
+
"learning_rate": 2.284263959390863e-06,
|
| 278 |
+
"loss": 0.637,
|
| 279 |
+
"step": 420
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 2.93,
|
| 283 |
+
"learning_rate": 1.015228426395939e-06,
|
| 284 |
+
"loss": 0.5974,
|
| 285 |
+
"step": 430
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 2.98,
|
| 289 |
+
"eval_accuracy": 0.8218390804597702,
|
| 290 |
+
"eval_loss": 0.5200982093811035,
|
| 291 |
+
"eval_runtime": 140.4463,
|
| 292 |
+
"eval_samples_per_second": 3.717,
|
| 293 |
+
"eval_steps_per_second": 0.47,
|
| 294 |
+
"step": 438
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 2.98,
|
| 298 |
+
"step": 438,
|
| 299 |
+
"total_flos": 3.483891679221596e+17,
|
| 300 |
+
"train_loss": 0.9891002352379229,
|
| 301 |
+
"train_runtime": 12153.7671,
|
| 302 |
+
"train_samples_per_second": 1.159,
|
| 303 |
+
"train_steps_per_second": 0.036
|
| 304 |
+
}
|
| 305 |
+
],
|
| 306 |
+
"logging_steps": 10,
|
| 307 |
+
"max_steps": 438,
|
| 308 |
+
"num_train_epochs": 3,
|
| 309 |
+
"save_steps": 500,
|
| 310 |
+
"total_flos": 3.483891679221596e+17,
|
| 311 |
+
"trial_name": null,
|
| 312 |
+
"trial_params": null
|
| 313 |
+
}
|