End of training
Browse files
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.98,
|
| 3 |
+
"eval_accuracy": 0.951271186440678,
|
| 4 |
+
"eval_loss": 0.13894420862197876,
|
| 5 |
+
"eval_runtime": 7.3751,
|
| 6 |
+
"eval_samples_per_second": 63.999,
|
| 7 |
+
"eval_steps_per_second": 8.0,
|
| 8 |
+
"total_flos": 3.1483475349026e+17,
|
| 9 |
+
"train_loss": 0.3312033372696,
|
| 10 |
+
"train_runtime": 688.7831,
|
| 11 |
+
"train_samples_per_second": 18.489,
|
| 12 |
+
"train_steps_per_second": 0.575
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.98,
|
| 3 |
+
"eval_accuracy": 0.951271186440678,
|
| 4 |
+
"eval_loss": 0.13894420862197876,
|
| 5 |
+
"eval_runtime": 7.3751,
|
| 6 |
+
"eval_samples_per_second": 63.999,
|
| 7 |
+
"eval_steps_per_second": 8.0
|
| 8 |
+
}
|
runs/Apr16_21-06-23_3901320cb440/events.out.tfevents.1681679941.3901320cb440.265.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:274198f0b2c195d42c0afefa70cd1a90b899c68cf1ce66822e4e704e51f89aee
|
| 3 |
+
size 411
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.98,
|
| 3 |
+
"total_flos": 3.1483475349026e+17,
|
| 4 |
+
"train_loss": 0.3312033372696,
|
| 5 |
+
"train_runtime": 688.7831,
|
| 6 |
+
"train_samples_per_second": 18.489,
|
| 7 |
+
"train_steps_per_second": 0.575
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,286 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.951271186440678,
|
| 3 |
+
"best_model_checkpoint": "swin-tiny-patch4-window7-224-bottom/checkpoint-265",
|
| 4 |
+
"epoch": 2.983050847457627,
|
| 5 |
+
"global_step": 396,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.08,
|
| 12 |
+
"learning_rate": 1.25e-05,
|
| 13 |
+
"loss": 1.0475,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.15,
|
| 18 |
+
"learning_rate": 2.5e-05,
|
| 19 |
+
"loss": 0.9521,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.23,
|
| 24 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 25 |
+
"loss": 0.7274,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.3,
|
| 30 |
+
"learning_rate": 5e-05,
|
| 31 |
+
"loss": 0.5366,
|
| 32 |
+
"step": 40
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.38,
|
| 36 |
+
"learning_rate": 4.859550561797753e-05,
|
| 37 |
+
"loss": 0.4714,
|
| 38 |
+
"step": 50
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.45,
|
| 42 |
+
"learning_rate": 4.719101123595506e-05,
|
| 43 |
+
"loss": 0.4572,
|
| 44 |
+
"step": 60
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.53,
|
| 48 |
+
"learning_rate": 4.578651685393259e-05,
|
| 49 |
+
"loss": 0.4177,
|
| 50 |
+
"step": 70
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.6,
|
| 54 |
+
"learning_rate": 4.438202247191011e-05,
|
| 55 |
+
"loss": 0.4239,
|
| 56 |
+
"step": 80
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.68,
|
| 60 |
+
"learning_rate": 4.297752808988764e-05,
|
| 61 |
+
"loss": 0.3931,
|
| 62 |
+
"step": 90
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.75,
|
| 66 |
+
"learning_rate": 4.157303370786517e-05,
|
| 67 |
+
"loss": 0.329,
|
| 68 |
+
"step": 100
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.83,
|
| 72 |
+
"learning_rate": 4.01685393258427e-05,
|
| 73 |
+
"loss": 0.3751,
|
| 74 |
+
"step": 110
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.9,
|
| 78 |
+
"learning_rate": 3.876404494382023e-05,
|
| 79 |
+
"loss": 0.357,
|
| 80 |
+
"step": 120
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.98,
|
| 84 |
+
"learning_rate": 3.735955056179776e-05,
|
| 85 |
+
"loss": 0.3578,
|
| 86 |
+
"step": 130
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.99,
|
| 90 |
+
"eval_accuracy": 0.9258474576271186,
|
| 91 |
+
"eval_loss": 0.19461879134178162,
|
| 92 |
+
"eval_runtime": 42.1119,
|
| 93 |
+
"eval_samples_per_second": 11.208,
|
| 94 |
+
"eval_steps_per_second": 1.401,
|
| 95 |
+
"step": 132
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 1.05,
|
| 99 |
+
"learning_rate": 3.595505617977528e-05,
|
| 100 |
+
"loss": 0.3842,
|
| 101 |
+
"step": 140
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 1.13,
|
| 105 |
+
"learning_rate": 3.455056179775281e-05,
|
| 106 |
+
"loss": 0.2794,
|
| 107 |
+
"step": 150
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 1.21,
|
| 111 |
+
"learning_rate": 3.314606741573034e-05,
|
| 112 |
+
"loss": 0.2737,
|
| 113 |
+
"step": 160
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 1.28,
|
| 117 |
+
"learning_rate": 3.1741573033707866e-05,
|
| 118 |
+
"loss": 0.1931,
|
| 119 |
+
"step": 170
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 1.36,
|
| 123 |
+
"learning_rate": 3.0337078651685396e-05,
|
| 124 |
+
"loss": 0.3022,
|
| 125 |
+
"step": 180
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 1.43,
|
| 129 |
+
"learning_rate": 2.893258426966292e-05,
|
| 130 |
+
"loss": 0.3115,
|
| 131 |
+
"step": 190
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 1.51,
|
| 135 |
+
"learning_rate": 2.752808988764045e-05,
|
| 136 |
+
"loss": 0.2525,
|
| 137 |
+
"step": 200
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 1.58,
|
| 141 |
+
"learning_rate": 2.6123595505617983e-05,
|
| 142 |
+
"loss": 0.2836,
|
| 143 |
+
"step": 210
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 1.66,
|
| 147 |
+
"learning_rate": 2.4719101123595505e-05,
|
| 148 |
+
"loss": 0.1842,
|
| 149 |
+
"step": 220
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 1.73,
|
| 153 |
+
"learning_rate": 2.3314606741573034e-05,
|
| 154 |
+
"loss": 0.2198,
|
| 155 |
+
"step": 230
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 1.81,
|
| 159 |
+
"learning_rate": 2.1910112359550563e-05,
|
| 160 |
+
"loss": 0.2833,
|
| 161 |
+
"step": 240
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 1.88,
|
| 165 |
+
"learning_rate": 2.0505617977528092e-05,
|
| 166 |
+
"loss": 0.2721,
|
| 167 |
+
"step": 250
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 1.96,
|
| 171 |
+
"learning_rate": 1.9101123595505618e-05,
|
| 172 |
+
"loss": 0.2537,
|
| 173 |
+
"step": 260
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 2.0,
|
| 177 |
+
"eval_accuracy": 0.951271186440678,
|
| 178 |
+
"eval_loss": 0.13894420862197876,
|
| 179 |
+
"eval_runtime": 7.2945,
|
| 180 |
+
"eval_samples_per_second": 64.706,
|
| 181 |
+
"eval_steps_per_second": 8.088,
|
| 182 |
+
"step": 265
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 2.03,
|
| 186 |
+
"learning_rate": 1.7696629213483147e-05,
|
| 187 |
+
"loss": 0.2686,
|
| 188 |
+
"step": 270
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 2.11,
|
| 192 |
+
"learning_rate": 1.6292134831460676e-05,
|
| 193 |
+
"loss": 0.222,
|
| 194 |
+
"step": 280
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 2.18,
|
| 198 |
+
"learning_rate": 1.4887640449438203e-05,
|
| 199 |
+
"loss": 0.2019,
|
| 200 |
+
"step": 290
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 2.26,
|
| 204 |
+
"learning_rate": 1.348314606741573e-05,
|
| 205 |
+
"loss": 0.2293,
|
| 206 |
+
"step": 300
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 2.34,
|
| 210 |
+
"learning_rate": 1.207865168539326e-05,
|
| 211 |
+
"loss": 0.1898,
|
| 212 |
+
"step": 310
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 2.41,
|
| 216 |
+
"learning_rate": 1.0674157303370787e-05,
|
| 217 |
+
"loss": 0.2073,
|
| 218 |
+
"step": 320
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 2.49,
|
| 222 |
+
"learning_rate": 9.269662921348316e-06,
|
| 223 |
+
"loss": 0.2081,
|
| 224 |
+
"step": 330
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 2.56,
|
| 228 |
+
"learning_rate": 7.865168539325843e-06,
|
| 229 |
+
"loss": 0.1857,
|
| 230 |
+
"step": 340
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 2.64,
|
| 234 |
+
"learning_rate": 6.460674157303372e-06,
|
| 235 |
+
"loss": 0.2429,
|
| 236 |
+
"step": 350
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 2.71,
|
| 240 |
+
"learning_rate": 5.056179775280899e-06,
|
| 241 |
+
"loss": 0.2008,
|
| 242 |
+
"step": 360
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 2.79,
|
| 246 |
+
"learning_rate": 3.651685393258427e-06,
|
| 247 |
+
"loss": 0.162,
|
| 248 |
+
"step": 370
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 2.86,
|
| 252 |
+
"learning_rate": 2.247191011235955e-06,
|
| 253 |
+
"loss": 0.159,
|
| 254 |
+
"step": 380
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 2.94,
|
| 258 |
+
"learning_rate": 8.426966292134832e-07,
|
| 259 |
+
"loss": 0.1909,
|
| 260 |
+
"step": 390
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 2.98,
|
| 264 |
+
"eval_accuracy": 0.951271186440678,
|
| 265 |
+
"eval_loss": 0.12845557928085327,
|
| 266 |
+
"eval_runtime": 6.8226,
|
| 267 |
+
"eval_samples_per_second": 69.182,
|
| 268 |
+
"eval_steps_per_second": 8.648,
|
| 269 |
+
"step": 396
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 2.98,
|
| 273 |
+
"step": 396,
|
| 274 |
+
"total_flos": 3.1483475349026e+17,
|
| 275 |
+
"train_loss": 0.3312033372696,
|
| 276 |
+
"train_runtime": 688.7831,
|
| 277 |
+
"train_samples_per_second": 18.489,
|
| 278 |
+
"train_steps_per_second": 0.575
|
| 279 |
+
}
|
| 280 |
+
],
|
| 281 |
+
"max_steps": 396,
|
| 282 |
+
"num_train_epochs": 3,
|
| 283 |
+
"total_flos": 3.1483475349026e+17,
|
| 284 |
+
"trial_name": null,
|
| 285 |
+
"trial_params": null
|
| 286 |
+
}
|