Table 3 Performance comparison for different models on the independent testing dataset.
From: Prediction of loess collapsibility coefficient using bayesian optimized random forest model
Regressor | R2 | EVS | MAE | RMSE | VAF | PI | RSR | NMBE |
|---|---|---|---|---|---|---|---|---|
DecisionTree | 0.884 | 0.886 | 0.012 | 0.017 | 88.6 | 1.753 | 0.97 | 0.006 |
Ridge | 0.434 | 0.462 | 0.03 | 0.039 | 46.212 | 0.858 | 2.143 | 0.031 |
RandomForest | 0.965 | 0.966 | 0.006 | 0.01 | 96.609 | 1.922 | 0.53 | 0.002 |
SVR | 0.897 | 0.908 | 0.014 | 0.016 | 90.806 | 1.789 | 0.914 | 0.006 |
LGBM | 0.961 | 0.961 | 0.008 | 0.01 | 96.059 | 1.911 | 0.565 | 0.002 |
xGBoost | 0.963 | 0.963 | 0.008 | 0.01 | 96.335 | 1.917 | 0.547 | 0.002 |
