几乎就像编写伪代码一样编写它。
In [14]: s = mn.groupby(['keyA','state'], as_index=False).sum()In [15]: s['v1'] = s['n1'] / s['d1']In [16]: s['v2'] = s['n2'] / s['d2']In [17]: s[['keyA', 'state', 'v1', 'v2']]Out[17]: keyA state v1v20 key1 AL 100 500.0000001 key1 CA 100 500.0000002 key1 NY 100 1500.0000003 key2 CA 100 1166.6666674 key2 FL 100 1166.666667[5 rows x 4 columns]
顺便说一句，我认为您的示例数据中有错字。第二个
n1标头应为
n2。
