preprocessing.py

# Used during prepprocessing to fix the sugar content values 
# that are incorrectly represented as "reg" instead of "Regular" in the dataset
def fix_sugar_content(X):
    X = X.copy()
    X["Product_Sugar_Content"] = X["Product_Sugar_Content"].replace("reg", "Regular")
    return X

# Feature engineering function to create a new feature for the 
# product ID prefix contained in the first two characters of the Product_Id
def add_product_id_prefix(X):
    X = X.copy()
    X["Product_Id_Prefix"] = X["Product_Id"].str[:2]
    return X