Table 4 LightGBM algorithm parameter setting.
From: A method for filling traffic data based on feature-based combination prediction model
Parameter name | Parameter meaning | Parameter range | Final selection value |
|---|---|---|---|
learning_rate | Learning rate | 0.05–0.2 | 0.1 |
n_estimators | Boosting iterations | 100–2000 | 1600 |
max_depth | Maximum depth of the tree model | 3–10 | 7 |
min_data_in_leaf | The minimum amount of data on a leaf | 20 | 20 |
min_child_weight | Minimum leaf node sample weight sum | 0.001 | 0.001 |
