Tune XGBoost Hyperparameters Systematically for Maximum Model Performance

You are an expert machine learning engineer with extensive experience tuning gradient boosting models. Guide me through a systematic XGBoost hyperparameter tuning process for my specific problem. **Problem Setup:** - Task type: [BINARY_CLASSIFICATION/MULTICLASS/REGRESSION] - Dataset size: [NUM_ROWS] rows × [NUM_FEATURES] features - Class imbalance ratio (if applicable): [IMBALANCE_RATIO] - Evaluation metric: [METRIC_NAME] - Computational budget: [LOW/MEDIUM/HIGH] (approximate time: [TIME_AVAILABLE]) - Current baseline performance: [BASELINE_SCORE] - Known data characteristics: [SPARSE_FEATURES/CATEGORICAL_HEAVY/HIGH_DIMENSIONAL/etc.] **Please provide a complete tuning strategy:** 1. **Phase 1 — Fix Learning Rate & Estimators**: Recommend an initial learning rate and use early stopping to find the optimal `n_estimators`. Provide the exact code snippet for this step. 2. **Phase 2 — Tree-Specific Parameters**: Define the search space and tuning order for `max_depth`, `min_child_weight`, and `gamma`. Explain WHY this order matters and provide recommended ranges based on my dataset size. 3. **Phase 3 — Regularization Parameters**: Guide me through tuning `subsample`, `colsample_bytree`, `reg_alpha` (L1), and `reg_lambda` (L2). Explain the interaction effects between these parameters. 4. **Phase 4 — Final Learning Rate Reduction**: Explain the technique of reducing the learning rate and proportionally increasing `n_estimators` for final performance gains. 5. **Search Strategy Recommendation**: Based on my computational budget, recommend the optimal search method (grid search, random search, Bayesian optimization with Optuna/Hyperopt) and provide a ready-to-run code template. 6. **Imbalance Handling** (if applicable): Recommend settings for `scale_pos_weight` or custom sample weights, and explain how this interacts with the tuning process. 7. **Overfitting Diagnostic Checklist**: Provide 5 specific signs that my XGBoost model is overfitting and the corresponding parameter adjustment for each. 8. **Final Configuration Template**: Output a complete, production-ready parameter dictionary with comments explaining each choice.