def process_data()

in blog_example_code/preprocessing.py [0:0]

• 8 lines of code
• 1 McCabe index (conditional complexity)

def process_data(df):
    df["sentiment"] = df["sentiment"].apply(labels_dict.get)
    df["review"] = df["review"].apply(preprocess_text)

    df["document"] = df[df.columns[::-1]].apply(
        lambda x: " ".join(x.dropna().astype(str)), axis=1
    )

    df.drop(["sentiment", "review"], axis=1, inplace=True)

    return df