Combining different models into a voting classifier

So far, we saw how to combine different instances of the same classifier or regressor into an ensemble. In this chapter, we are going to take this idea a step further and combine conceptually different classifiers into what is known as a voting classifier.

The idea behind voting classifiers is that the individual learners in the ensemble don't necessarily need to be of the same type. After all, no matter how the individual classifiers arrived at their prediction, in the end, we are going to apply a decision rule that integrates all the votes of the individual classifiers. This is also known as a voting scheme.

Understanding different voting schemes

Two different voting schemes are common among voting classifiers:

• In hard voting (also known as majority voting), every individual classifier votes for a class, and the majority wins. In statistical terms, the predicted target label of the ensemble is the mode of the distribution of individually...