Table 4 Results of hyperparameter optimization using XGBoost
From: Comparison of factors influencing landslide risk near a forest road in Chungju-si, South Korea
Train and test data ratio | 9:1 | 8:2 | 7:3 | 6:4 | 5:5 |
|---|---|---|---|---|---|
learning_rate | 0.01 | 0.01 | 0.01 | 0.04 | 0.02 |
n_estimators | 500 | 500 | 500 | 3 | 2 |
max_depth | 3 | 3 | 3 | 500 | 500 |
gamma | 1 | 2 | 2 | 1 | 1 |
colsample_bytree | 0.5 | 1 | 1 | 0 | 0 |
max_delta_step | 1 | 0 | 0 | 4 | 3.0 |
min_child_weight | 1 | 0.5 | 0.5 | 0.5 | 1.5 |
reg_alpha | 1 | 0 | 0 | 1.5 | 0 |
reg_lambda | 2.5 | 2.5 | 2.5 | 2.5 | 0.5 |
subsample | 1 | 1 | 1 | 1 | 1 |
scale_pos_weight | 1 | 1 | 1 | 2 | 1 |