Encode categorical and scale numerical variables
In this final step, you will perform one-hot encoding on the categorical variables and then scale the numerical columns. The pandas library has been loaded for you as pd, as well as the StandardScaler module from the sklearn.preprocessing module.
The raw telecom churn dataset telco_raw has been loaded for you as a pandas DataFrame, as well as the lists custid, target, categorical, and numerical with column names you have created in the previous exercise. You can familiarize yourself with the dataset by exploring it in the console.
Cet exercice fait partie du cours
Machine Learning for Marketing in Python
Instructions
- Perform one-hot encoding on the categorical variables.
- Initialize a
StandardScalerinstance. - Fit and transform the
scaleron the numerical columns. - Build a DataFrame from
scaled_numerical.
Exercice interactif pratique
Essayez cet exercice en complétant cet exemple de code.
# Perform one-hot encoding to categorical variables
telco_raw = pd.get_dummies(data = ___, columns = categorical, drop_first=True)
# Initialize StandardScaler instance
scaler = ___()
# Fit and transform the scaler on numerical columns
scaled_numerical = ___.fit_transform(telco_raw[___])
# Build a DataFrame from scaled_numerical
scaled_numerical = pd.DataFrame(___, columns=numerical)