def topn_values(column_name, n):
    """
    Take a column name and find the most frequent values
    """
    return df.groupby(column_name).count().sort('count', ascending=False).limit(n).toPandas()