Introduction
In the previous chapter, we learned about a utility function called the ML pipeline, which automates various processes, such as scaling, dimensionality reduction, and modeling, within the data science life cycle.
In this chapter, we will learn about another utility that helps in automating feature engineering. We have completed different feature engineering tasks in the previous chapters, such as in Chapter 3, Binary Classification, and Chapter 12, Feature Engineering. When building features in the previous exercises, you would have realized how tedious this step is when it's done manually.
For instance, in Chapter 3, Binary Classification, in Exercise 3.02, you implemented different aggregation functions using the traditional manual ways to create a new feature, as shown in the following code snippet:
# Aggregation on age
ageTot = bankData.groupby('age')['y'].agg(ageTot='count').reset_index()
ageTot.head()
