Research Article
Coupling a Bat Algorithm with XGBoost to Estimate Reference Evapotranspiration in the Arid and Semiarid Regions of China
Table 4
Results of the statistical indicators for all models in the Datong station during the training and testing stages.
| Model | Training | Testing | R2 | RMSE | MAE | RMSEmax10% | MAEmax10% | R2 | RMSE | MAE | RMSEmax10% | MAEmax10% |
| GPR1 | 0.841 | 0.66 | 0.519 | 1.392 | 1.373 | 0.832 | 0.695 | 0.559 | 1.333 | 1.328 | GPR2 | 0.978 | 0.258 | 0.196 | 0.586 | 0.563 | 0.972 | 0.372 | 0.302 | 0.744 | 0.738 | GPR3 | 0.98 | 0.242 | 0.181 | 0.565 | 0.547 | 0.956 | 0.444 | 0.35 | 0.898 | 0.891 | GPR4 | 0.974 | 0.279 | 0.202 | 0.673 | 0.645 | 0.978 | 0.297 | 0.221 | 0.668 | 0.664 | GPR5 | 0.988 | 0.197 | 0.138 | 0.483 | 0.46 | 0.951 | 0.398 | 0.251 | 1.089 | 1.019 | GPR6 | 0.986 | 0.209 | 0.139 | 0.548 | 0.517 | 0.958 | 0.37 | 0.25 | 0.919 | 0.877 | MARS1 | 0.964 | 0.315 | 0.236 | 0.723 | 0.71 | 0.97 | 0.292 | 0.227 | 0.649 | 0.631 | MARS2 | 0.986 | 0.199 | 0.15 | 0.465 | 0.453 | 0.962 | 0.421 | 0.337 | 0.821 | 0.812 | MARS3 | 0.984 | 0.209 | 0.157 | 0.481 | 0.468 | 0.953 | 0.512 | 0.379 | 1.158 | 1.148 | MARS4 | 0.99 | 0.162 | 0.121 | 0.371 | 0.363 | 0.98 | 0.271 | 0.216 | 0.559 | 0.552 | MARS5 | 0.992 | 0.154 | 0.11 | 0.377 | 0.362 | 0.992 | 0.153 | 0.12 | 0.321 | 0.32 | MARS6 | 0.994 | 0.122 | 0.087 | 0.3 | 0.292 | 0.992 | 0.189 | 0.121 | 0.509 | 0.483 | BAXGB1 | 0.978 | 0.238 | 0.179 | 0.541 | 0.531 | 0.972 | 0.291 | 0.23 | 0.592 | 0.587 | BAXGB2 | 0.988 | 0.209 | 0.148 | 0.518 | 0.505 | 0.974 | 0.284 | 0.227 | 0.576 | 0.568 | BAXGB3 | 0.998 | 0.076 | 0.057 | 0.175 | 0.168 | 0.955 | 0.438 | 0.311 | 1.067 | 1.057 | BAXGB4 | 0.996 | 0.092 | 0.069 | 0.214 | 0.207 | 0.988 | 0.191 | 0.153 | 0.403 | 0.398 | BAXGB5 | 0.996 | 0.108 | 0.076 | 0.266 | 0.255 | 0.992 | 0.146 | 0.112 | 0.337 | 0.328 | BAXGB6 | 1 | 0.03 | 0.022 | 0.068 | 0.066 | 0.996 | 0.116 | 0.087 | 0.272 | 0.258 |
|
|
Note. Bold font indicates the best model.
|