Table 4 Optimization points for regression models.
From: Reliable and efficient solar radiation estimation with the insights of XAI
Model | Key parameters | Optimization strategies |
|---|---|---|
Decision tree | max_depth, min_samples_leaf, criterion | 1. Limit tree depth to reduce overfitting. |
2. Use grid search with cross-validation. | ||
Random forest | n_estimators, max_features, max_depth | 1. Increase n_estimators until OOB error stabilizes. |
2. Tune max_features for bias–variance control. | ||
AdaBoost | n_estimators, learning_rate | 1.Balance n_estimators and learning_rate. |
2. Use shallow base learners (depth 1–3). | ||
Gradient boosting | learning_rate, n_estimators, subsample | 1. Use small learning_rate with more estimators. |
2. Apply early stopping for generalization. | ||
XGBoost | eta, max_depth, subsample, lambda | 1. Tune regularization (lambda, alpha). |
2. Use early stopping and balanced learning rate. |
