我正在导入一个HTML文件。它有一个奇怪的格式和多个索引的数据。

我对导入表"Photovoltaic"特别感兴趣，它从大表的第10行开始。该表似乎是多索引的。
代码：

net_met_cus = 'https://www.eia.gov/electricity/annual/html/epa_04_10.html'
    
    net_met = pd.read_html(net_met_cus)
    print(len(net_met))
    net_met_pv = net_met[1]
    # Photovoltaic table starts at 12 row
    print(net_met_pv.loc[12])
Unnamed: 0_level_0  Year              Photovoltaic
Capacity (MW)       Residential       Photovoltaic
                    Commercial        Photovoltaic
                    Industrial        Photovoltaic
                    Transportation    Photovoltaic
                    Total             Photovoltaic
Customers           Residential       Photovoltaic
                    Commercial        Photovoltaic
                    Industrial        Photovoltaic
                    Transportation    Photovoltaic
                    Total             Photovoltaic
Name: 12, dtype: object
    # Is it multiindex
    print(net_met_pv.loc[12].index)
MultiIndex([('Unnamed: 0_level_0',           'Year'),
            (     'Capacity (MW)',    'Residential'),
            (     'Capacity (MW)',     'Commercial'),
            (     'Capacity (MW)',     'Industrial'),
            (     'Capacity (MW)', 'Transportation'),
            (     'Capacity (MW)',          'Total'),
            (         'Customers',    'Residential'),
            (         'Customers',     'Commercial'),
            (         'Customers',     'Industrial'),
            (         'Customers', 'Transportation'),
            (         'Customers',          'Total')],
           )
    # Okay, let's flaten it
    net_met_pv.to_flat_index()

当前输出：

AttributeError: 'DataFrame' object has no attribute 'to_flat_index'