Hands-On Data Analysis with Pandas

Logistic regression

Logistic regression is a way to use linear regression to solve classification tasks. However, it uses the logistic sigmoid function to return probabilities in [0, 1] that can be mapped to class labels:

Let's use logistic regression to classify red wines as high or low quality and to classify wines as red or white based on their chemical properties. We can treat logistic regression as we did the linear regression in the previous section, using the linear_model module in scikit-learn. Just like the linear regression problem, we will be using a supervised method, so we have to split our data into testing and training sets. ...