Learning Data Mining with Python - Second Edition

Standard pre-processing

The pre-processing we will perform for this experiment is called feature-based normalization, which we perform using scikit-learn's MinMaxScaler class. Continuing with the Jupyter Notebook from the rest of this chapter, first, we import this class:

fromsklearn.preprocessing import MinMaxScaler

This class takes each feature and scales it to the range 0 to 1. This pre-processor replaces the minimum value with 0, the maximum with 1, and the other values somewhere in between based on a linear mapping.

To apply our pre-processor, we run the transform function on it. Transformers often need to be trained first, in the same way that the classifiers do. We can combine these steps by running the fit_transform function instead: ...