This project is designed to practice various data science concepts, including:
Training and tuning classification models
Performing feature engineering to improve model performance
Explaining, interpreting, and debugging models
Model: Train and tune two types of models: GBM (H2O) and LightGBM.
Scoring: Single scoring function for either GBM or LightGBM
Interpretation: Provided a detailed write-up on what features are important for
model predictions (Shapley values)
コメント