Summary
In this chapter, you learned some of the well-tested tips and tricks from the winners of Kaggle competitions. In addition to exploring Kaggle competitions and understanding the importance of a hold-out set, you gained essential practice in feature engineering time columns, feature engineering categorical columns, mean encoding, building non-correlated ensembles, and stacking. These advanced techniques are widespread among elite Kagglers, and they can give you an edge when developing machine learning models for research, competition, and industry.
In the next and final chapter, we will shift gears from the competitive world to the tech world, where we will build an XGBoost model from beginning to end using transformers and pipelines to complete a model ready for industry deployment.
