Python Feature Engineering Cookbook

There's more...

Let's visualize the relationship of the polynomial variables with the target. First, let's run the main recipe as indicated in step 1 to step 8. Then, let's import the Python visualization library and make the plots:

1. Import Matplotlib:

import matplotlib.pyplot as plt

2. Let's create a function to make multiple subplots, each displaying one of the new polynomial features in a scatter plot versus the target:

def plot_features(df, target):    nb_rows = 5    nb_cols = 4        fig, axs = plt.subplots(nb_rows, nb_cols, figsize=(12, 12))    plt.subplots_adjust(wspace=None, hspace=0.4)    n = 0    for i in range(0, nb_rows):        for j in range(0, nb_cols):            if n!=19:                axs[i, j].scatter(df[df.columns[n]], target)                axs[i, j].set_title(df.columns[n])                n += 1