encode_values.py

encode_values.py

from sklearn.preprocessing import OneHotEncoder

categorical_columns = ['Pclass', 'Sex', 'Embarked', 'cabin_letter']
categorical_encoder = OneHotEncoder(handle_unknown='ignore')
categorical_encoder.fit(df[categorical_columns])

# Add the new columns to the data
new_column_names = []
for idx, cat_column_name in enumerate(categorical_columns):
    values = categorical_encoder.categories_[idx]
    new_column_names += [f'{cat_column_name}_{value}' for value in values]

df.loc[:, new_column_names] = \
  categorical_encoder.transform(df[categorical_columns]).toarray()