比较具有相同列名的两个数据帧,并用后缀展平生成的数据帧

我有两个数据帧:

dfA = pd.DataFrame({
        "Run": (
            1, 1, 1, 2, 2, 2, 3, 3, 3
        ),
        "Point": (
            1, 2, 3, 1, 2, 3, 1, 2, 3
        ),
        "Val": (
            78, 79, 77, 78, 79, 77, 78, 79, 77
        )
    })

和

dfB = pd.DataFrame({
        "Run": (
            1, 1, 1, 2, 2, 2, 3, 3, 3,
        ),
        "Point": (
            1, 2, 3, 1, 2, 3, 1, 2, 3, 
        ),
        "Val": (
            68, 69, 67, 68, 69, 67, 68, 69, 67,
        ),
    })

具有

res = dfA.compare(dfB, keep_equal=True, keep_shape=True).rename(columns={'self': 'A', 'other': 'B'}, level=-1)

我得到了一个与我想要的结果完全一致的数据帧,但Dash AG Grid似乎无法处理多索引列数据帧,因此我想将“res”df压平以获得以下结构

end_df = pd.DataFrame({
        "Run_A": (
            1, 1, 1, 2, 2, 2, 3, 3, 3,
        ),
        "Run_B": (
            1, 1, 1, 2, 2, 2, 3, 3, 3,
        ),
        "Point_A": (
            1, 2, 3, 1, 2, 3, 1, 2, 3, 
        ),
        "Point_B": (
            1, 2, 3, 1, 2, 3, 1, 2, 3, 
        ),
        "Val_A": (
            78, 79, 77, 78, 79, 77, 78, 79, 77
        ),
        "Val_B": (
            68, 69, 67, 68, 69, 67, 68, 69, 67,
        ),
    })

归档所需结构的最有效方法是什么?