Commit
•
da48b3f
1
Parent(s):
dd3c386
End of training
Browse files- all_results.json +13 -0
- eval_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +295 -0
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_accuracy": 0.9709876543209877,
|
| 4 |
+
"eval_loss": 0.10647899657487869,
|
| 5 |
+
"eval_runtime": 7.1928,
|
| 6 |
+
"eval_samples_per_second": 225.226,
|
| 7 |
+
"eval_steps_per_second": 7.09,
|
| 8 |
+
"total_flos": 1.0874439903456461e+18,
|
| 9 |
+
"train_loss": 0.4010893177916432,
|
| 10 |
+
"train_runtime": 447.7729,
|
| 11 |
+
"train_samples_per_second": 97.683,
|
| 12 |
+
"train_steps_per_second": 0.764
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_accuracy": 0.9709876543209877,
|
| 4 |
+
"eval_loss": 0.10647899657487869,
|
| 5 |
+
"eval_runtime": 7.1928,
|
| 6 |
+
"eval_samples_per_second": 225.226,
|
| 7 |
+
"eval_steps_per_second": 7.09
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 1.0874439903456461e+18,
|
| 4 |
+
"train_loss": 0.4010893177916432,
|
| 5 |
+
"train_runtime": 447.7729,
|
| 6 |
+
"train_samples_per_second": 97.683,
|
| 7 |
+
"train_steps_per_second": 0.764
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,295 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9709876543209877,
|
| 3 |
+
"best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat\\checkpoint-342",
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 342,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.09,
|
| 13 |
+
"grad_norm": 5.281110763549805,
|
| 14 |
+
"learning_rate": 1.4285714285714285e-05,
|
| 15 |
+
"loss": 2.2704,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.18,
|
| 20 |
+
"grad_norm": 6.057841777801514,
|
| 21 |
+
"learning_rate": 2.857142857142857e-05,
|
| 22 |
+
"loss": 1.9317,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.26,
|
| 27 |
+
"grad_norm": 17.515657424926758,
|
| 28 |
+
"learning_rate": 4.2857142857142856e-05,
|
| 29 |
+
"loss": 1.272,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.35,
|
| 34 |
+
"grad_norm": 23.613327026367188,
|
| 35 |
+
"learning_rate": 4.9185667752443e-05,
|
| 36 |
+
"loss": 0.6814,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.44,
|
| 41 |
+
"grad_norm": 18.322710037231445,
|
| 42 |
+
"learning_rate": 4.755700325732899e-05,
|
| 43 |
+
"loss": 0.5119,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.53,
|
| 48 |
+
"grad_norm": 25.08259391784668,
|
| 49 |
+
"learning_rate": 4.592833876221499e-05,
|
| 50 |
+
"loss": 0.4619,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.61,
|
| 55 |
+
"grad_norm": 16.223060607910156,
|
| 56 |
+
"learning_rate": 4.429967426710098e-05,
|
| 57 |
+
"loss": 0.4132,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.7,
|
| 62 |
+
"grad_norm": 16.03809928894043,
|
| 63 |
+
"learning_rate": 4.2671009771986977e-05,
|
| 64 |
+
"loss": 0.4062,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.79,
|
| 69 |
+
"grad_norm": 10.12449836730957,
|
| 70 |
+
"learning_rate": 4.104234527687297e-05,
|
| 71 |
+
"loss": 0.3142,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.88,
|
| 76 |
+
"grad_norm": 9.619889259338379,
|
| 77 |
+
"learning_rate": 3.941368078175896e-05,
|
| 78 |
+
"loss": 0.3146,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.96,
|
| 83 |
+
"grad_norm": 11.356223106384277,
|
| 84 |
+
"learning_rate": 3.778501628664495e-05,
|
| 85 |
+
"loss": 0.2841,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 1.0,
|
| 90 |
+
"eval_accuracy": 0.9469135802469136,
|
| 91 |
+
"eval_loss": 0.17170849442481995,
|
| 92 |
+
"eval_runtime": 6.6951,
|
| 93 |
+
"eval_samples_per_second": 241.968,
|
| 94 |
+
"eval_steps_per_second": 7.618,
|
| 95 |
+
"step": 114
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 1.05,
|
| 99 |
+
"grad_norm": 13.724865913391113,
|
| 100 |
+
"learning_rate": 3.615635179153095e-05,
|
| 101 |
+
"loss": 0.308,
|
| 102 |
+
"step": 120
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.14,
|
| 106 |
+
"grad_norm": 18.163928985595703,
|
| 107 |
+
"learning_rate": 3.452768729641694e-05,
|
| 108 |
+
"loss": 0.2901,
|
| 109 |
+
"step": 130
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 1.23,
|
| 113 |
+
"grad_norm": 10.404041290283203,
|
| 114 |
+
"learning_rate": 3.289902280130293e-05,
|
| 115 |
+
"loss": 0.2516,
|
| 116 |
+
"step": 140
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 1.32,
|
| 120 |
+
"grad_norm": 24.62061309814453,
|
| 121 |
+
"learning_rate": 3.127035830618892e-05,
|
| 122 |
+
"loss": 0.2744,
|
| 123 |
+
"step": 150
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 1.4,
|
| 127 |
+
"grad_norm": 26.786832809448242,
|
| 128 |
+
"learning_rate": 2.9641693811074923e-05,
|
| 129 |
+
"loss": 0.24,
|
| 130 |
+
"step": 160
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 1.49,
|
| 134 |
+
"grad_norm": 11.213741302490234,
|
| 135 |
+
"learning_rate": 2.8013029315960915e-05,
|
| 136 |
+
"loss": 0.2152,
|
| 137 |
+
"step": 170
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 1.58,
|
| 141 |
+
"grad_norm": 14.150710105895996,
|
| 142 |
+
"learning_rate": 2.6384364820846906e-05,
|
| 143 |
+
"loss": 0.2344,
|
| 144 |
+
"step": 180
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 1.67,
|
| 148 |
+
"grad_norm": 7.0177812576293945,
|
| 149 |
+
"learning_rate": 2.47557003257329e-05,
|
| 150 |
+
"loss": 0.2265,
|
| 151 |
+
"step": 190
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 1.75,
|
| 155 |
+
"grad_norm": 13.405513763427734,
|
| 156 |
+
"learning_rate": 2.3127035830618892e-05,
|
| 157 |
+
"loss": 0.253,
|
| 158 |
+
"step": 200
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 1.84,
|
| 162 |
+
"grad_norm": 7.144536018371582,
|
| 163 |
+
"learning_rate": 2.149837133550489e-05,
|
| 164 |
+
"loss": 0.2086,
|
| 165 |
+
"step": 210
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 1.93,
|
| 169 |
+
"grad_norm": 17.144977569580078,
|
| 170 |
+
"learning_rate": 1.986970684039088e-05,
|
| 171 |
+
"loss": 0.2053,
|
| 172 |
+
"step": 220
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 2.0,
|
| 176 |
+
"eval_accuracy": 0.9561728395061728,
|
| 177 |
+
"eval_loss": 0.1471811980009079,
|
| 178 |
+
"eval_runtime": 6.8477,
|
| 179 |
+
"eval_samples_per_second": 236.576,
|
| 180 |
+
"eval_steps_per_second": 7.448,
|
| 181 |
+
"step": 228
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 2.02,
|
| 185 |
+
"grad_norm": 8.713410377502441,
|
| 186 |
+
"learning_rate": 1.8241042345276872e-05,
|
| 187 |
+
"loss": 0.2034,
|
| 188 |
+
"step": 230
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 2.11,
|
| 192 |
+
"grad_norm": 10.655760765075684,
|
| 193 |
+
"learning_rate": 1.6612377850162867e-05,
|
| 194 |
+
"loss": 0.1967,
|
| 195 |
+
"step": 240
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 2.19,
|
| 199 |
+
"grad_norm": 22.123258590698242,
|
| 200 |
+
"learning_rate": 1.4983713355048862e-05,
|
| 201 |
+
"loss": 0.1745,
|
| 202 |
+
"step": 250
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 2.28,
|
| 206 |
+
"grad_norm": 14.105958938598633,
|
| 207 |
+
"learning_rate": 1.3355048859934855e-05,
|
| 208 |
+
"loss": 0.207,
|
| 209 |
+
"step": 260
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 2.37,
|
| 213 |
+
"grad_norm": 11.178110122680664,
|
| 214 |
+
"learning_rate": 1.1726384364820847e-05,
|
| 215 |
+
"loss": 0.1848,
|
| 216 |
+
"step": 270
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 2.46,
|
| 220 |
+
"grad_norm": 8.699991226196289,
|
| 221 |
+
"learning_rate": 1.009771986970684e-05,
|
| 222 |
+
"loss": 0.1661,
|
| 223 |
+
"step": 280
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 2.54,
|
| 227 |
+
"grad_norm": 13.814720153808594,
|
| 228 |
+
"learning_rate": 8.469055374592833e-06,
|
| 229 |
+
"loss": 0.1564,
|
| 230 |
+
"step": 290
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 2.63,
|
| 234 |
+
"grad_norm": 7.040809631347656,
|
| 235 |
+
"learning_rate": 6.840390879478828e-06,
|
| 236 |
+
"loss": 0.15,
|
| 237 |
+
"step": 300
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 2.72,
|
| 241 |
+
"grad_norm": 11.934651374816895,
|
| 242 |
+
"learning_rate": 5.211726384364822e-06,
|
| 243 |
+
"loss": 0.1848,
|
| 244 |
+
"step": 310
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 2.81,
|
| 248 |
+
"grad_norm": 7.909034729003906,
|
| 249 |
+
"learning_rate": 3.5830618892508147e-06,
|
| 250 |
+
"loss": 0.1724,
|
| 251 |
+
"step": 320
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 2.89,
|
| 255 |
+
"grad_norm": 10.677735328674316,
|
| 256 |
+
"learning_rate": 1.9543973941368076e-06,
|
| 257 |
+
"loss": 0.1456,
|
| 258 |
+
"step": 330
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 2.98,
|
| 262 |
+
"grad_norm": 7.879863262176514,
|
| 263 |
+
"learning_rate": 3.2573289902280136e-07,
|
| 264 |
+
"loss": 0.166,
|
| 265 |
+
"step": 340
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 3.0,
|
| 269 |
+
"eval_accuracy": 0.9709876543209877,
|
| 270 |
+
"eval_loss": 0.10647899657487869,
|
| 271 |
+
"eval_runtime": 6.8337,
|
| 272 |
+
"eval_samples_per_second": 237.059,
|
| 273 |
+
"eval_steps_per_second": 7.463,
|
| 274 |
+
"step": 342
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 3.0,
|
| 278 |
+
"step": 342,
|
| 279 |
+
"total_flos": 1.0874439903456461e+18,
|
| 280 |
+
"train_loss": 0.4010893177916432,
|
| 281 |
+
"train_runtime": 447.7729,
|
| 282 |
+
"train_samples_per_second": 97.683,
|
| 283 |
+
"train_steps_per_second": 0.764
|
| 284 |
+
}
|
| 285 |
+
],
|
| 286 |
+
"logging_steps": 10,
|
| 287 |
+
"max_steps": 342,
|
| 288 |
+
"num_input_tokens_seen": 0,
|
| 289 |
+
"num_train_epochs": 3,
|
| 290 |
+
"save_steps": 500,
|
| 291 |
+
"total_flos": 1.0874439903456461e+18,
|
| 292 |
+
"train_batch_size": 32,
|
| 293 |
+
"trial_name": null,
|
| 294 |
+
"trial_params": null
|
| 295 |
+
}
|