Set difference of rows of two data frames

df_diff.py

def df_diff(d1, d2):
    """
    df_diff Create a DataFrame containing rows of d1 not in d2

    Arguments:
        d1 -- A data frame
        d2 -- Another DataFrame which is a subset of d1

    Returns:
        A pandas DataFrame containing rows of d1 that are not in d2
    """
    df_all = d1.merge(d2.drop_duplicates(), how="left", indicator=True)
    return df_all[df_all._merge == "left_only"].drop(columns="_merge")