基于机器学习的青藏高原高寒沼泽湿地蒸散发插补研究
|
王秀英, 陈奇, 杜华礼, 张睿, 马红璐
|
Evapotranspiration interpolation in alpine marshes wetland on the Qingzang Plateau based on machine learning
|
WANG Xiu-Ying, CHEN Qi, DU Hua-Li, ZHANG Rui, MA Hong-Lu
|
|
表4 隆宝高寒沼泽湿地不同气象因子组合下的模型精度
|
Table 4 Model accuracy under the combination of different meteorological factors in Longbao alpine marshes wetland
|
|
模型特征组合 Model feature combination | MLR | CART | RF | SVR | MLP | R2平均 Mean of R2 | R2 | RMSE | R2 | RMSE | R2 | RMSE | R2 | RMSE | R2 | RMSE | 组合1 Combination 1 | 0.81 | 0.041 | 0.72 | 0.051 | 0.85 | 0.039 | 0.77 | 0.056 | 0.80 | 0.053 | 0.79 | 组合2 Combination 2 | 0.80 | 0.043 | 0.72 | 0.052 | 0.82 | 0.041 | 0.68 | 0.063 | 0.69 | 0.062 | 0.74 | 组合3 Combination 3 | 0.82 | 0.041 | 0.72 | 0.052 | 0.83 | 0.040 | 0.72 | 0.073 | 0.58 | 0.063 | 0.73 | 组合4 Combination 4 | 0.80 | 0.042 | 0.72 | 0.052 | 0.83 | 0.041 | 0.69 | 0.079 | 0.66 | 0.066 | 0.74 | 组合5 Combination 5 | 0.77 | 0.063 | 0.68 | 0.057 | 0.82 | 0.042 | 0.78 | 0.072 | 0.55 | 0.129 | 0.72 | 组合6 Combination 6 | 0.71 | 0.045 | 0.71 | 0.043 | 0.82 | 0.031 | 0.69 | 0.093 | 0.36 | 0.136 | 0.66 | 组合7 Combination 7 | 0.57 | 0.223 | 0.71 | 0.044 | 0.83 | 0.033 | 0.35 | 0.118 | 0.41 | 0.117 | 0.57 | 平均 Mean | 0.75 | 0.071 | 0.71 | 0.050 | 0.83 | 0.038 | 0.67 | 0.079 | 0.58 | 0.089 | 0.71 |
|
|
|