Probability ratio encoding

Probability ratio encoding transforms categorical variables into numerical values by calculating the ratio of the probability of the target being 1 to the probability of it being 0 for each category. This encoding technique captures the distinct contribution of each category to the target outcome, providing valuable information for building effective models. Let’s get started:

1. To begin, import pandas and train_test_split from scikit-learn to prepare for applying the encoding. Import matplotlib.pyplot, which you will use to make graphs showing the relationship between the encoded categorical variables and the target variable:

   import pandas as pd
   from sklearn.model_selection import train_test_split
   import matplotlib.pyplot as plt

2. Next, set up the sample data. This sample dataset has two categorical columns, Neighborhood and Exterior1st, and a target column, SalePrice:

   # Sample data
   data = {
       'Neighborhood': ...