describe_categorical_values.py

def describe_categorical_values(df, non_interesting_columns=[], num_categories=5):
    values_df = pd.DataFrame()
    for i, column in enumerate(df.columns):
        if column in non_interesting_columns:
            continue
        top_values0 = ["{}: {}%".format(x,int(round(100*y/len(df))))
                       for x, y in zip(df[column].value_counts(dropna=False).head(num_categories).index,
                                       df[column].value_counts(dropna=False).head(num_categories).values)]
        if len(top_values0) < num_categories:
            top_values = [None]*num_categories
            top_values[:len(top_values0)] = top_values0
        else:
            top_values = top_values0
        values_df[column] = top_values
    return values_df.transpose()