Python Feature Engineering Cookbook

How it works...

In this recipe, we applied power transformations using NumPy, scikit-learn, and Feature-engine while using the Boston House Prices dataset from scikit-learn.

To apply exponential functions with NumPy, we created a copy of the original dataframe with pandas copy(). Next, we used the power() method on a slice of the dataset with the variables to transform and captured the transformed variables in the new dataframe. This procedure returned a pandas dataframe with the original variables, and LSTAT, NOX, DIS, and RM were transformed with a power of 0.3.

To apply an exponential transformation with scikit-learn, we used FunctionTransformer(), which applies a user-defined function. We started the transformer with np.power() within ...