Table 9.
Hyperparameters | Meanings | Search ranges | Optimal values |
---|---|---|---|
learning_rate | Learning rate | (0.01, 1.0) | 0.1 |
max_depth | Maximum depth of the tree | (1, 50) | 15 |
max_bin | The max number of bins that feature values will be bucketed in | (10, 100) | 20 |
reg_alpha | L1 regularization | (1e-9, 1.0) | 1e-5 |
boosting_type | Training method | gbdt; goss; rf; dart | gbdt |
num_leaves | Number of leaf nodes | (4, 50) | 36 |
n_estimators | Number of iterations | (100, 600) | 450 |