Hands-On Data Analysis with Pandas

Logistic regression

Since logistic regression is another simple model, let's try it out next. We used logistic regression in Chapter 9, Getting Started with Machine Learning in Python, for classification problems, so we already know how it works. As we learned in Chapter 10, Making Better Predictions – Optimizing Models, we will use a grid search to find a good value for the regularization parameter in our desired search space, using recall_macro for scoring. Remember there is a large cost associated with false negatives, so we are focusing on recall. The _macro suffix indicates that we want to average the recall between the positive and negative classes, instead of looking at it overall (due to the class imbalance).

If we know exactly how ...