We will use the following Python libraries: pandas, NumPy, Matplotlib, scikit-learn, and Feature-engine. In the Technical requirements section of Chapter 1, Foreseeing Variable Problems in Building ML Models, you will find instructions on how to install these libraries. To install Feature-engine, you can use pip: pip install feature-engine. Throughout the recipes, we will use the Boston House Prices dataset from scikit-learn, which contains no missing data.
To perform discretization in your own datasets, make sure you impute missing data with any of the techniques covered in Chapter 2, Imputing Missing Data.
