3

Regularization with Linear Models

A huge part of machine learning (ML) is made up of linear models. Although sometimes considered less powerful than their nonlinear counterparts (such as tree-based models or deep learning models), linear models do address many concrete, valuable problems. Customer churn and advertising optimization are just a couple of problems where linear models may be the right solution.

In this chapter, we will cover the following recipes:

  • Training a linear regression with scikit-learn
  • Regularizing with ridge regression
  • Regularizing with lasso regression
  • Regularizing with elastic net regression
  • Training a logistic regression model
  • Regularizing a logistic regression model
  • Choosing the right regularization

By the end of ...

Get The Regularization Cookbook now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.