| Parameter | Range | Used | Description |
| max_depth | [3, 4, 5, 6, 7, 8, 9, 10] | 3 | Maximum tree depth for base learners | min_child_weight | [1, 2, 3, 4, 5] | 4 | Minimum sum of instance weight (hessian) needed in a child | gamma | [0, 0.1, 0.2, 0.3, 0.4, 0.5] | 0.2 | Minimum loss reduction required to make a further partition on a leaf | subsample | [0.6, 0.7, 0.8, 0.9, 1] | 0.8 | Subsample ratio of the training instance | colsample_bytree | [0.6, 0.7, 0.8, 0.9, 1] | 0.8 | Subsample ratio of columns when constructing each tree | reg_alpha | [1e − 5, 1e − 4, 1e − 3, 1e − 2, 0.1, 1, 100] | 1e − 05 | L1 regularization term on weights | reg_lambda | [0.05, 0.1, 1, 2, 3] | 1 | L2 regularization term on weights | learning_rate | [0.01, 0.05, 0.07, 0.1, 0.2] | 0.07 | Boosting learning rate |
|
|