python - 基于 Pandas 列的分组和自动递增

Question

我有一个看起来像这样的 pandas 框架:

有没有一种方法可以在不必遍历数据框的情况下将数字添加到最后一列？

我正在玩 Grouping and auto incrementing group id in pandas 的结果但还没有让它为我的目的工作

这是生成数据框的代码

import pandas as pd
columns = ['Product','SubProd', 'NeedThis']
Index=['4/20/2012','4/27/2012','5/4/2012','5/11/2012','5/18/2012','4/20/2012',
'4/27/2012','5/4/2012','5/11/2012','5/18/2012','5/25/2012','10/31/2014','11/7/2014',
'11/14/2014','11/21/2014','11/28/2014']
datas = {'Product' : ['A','A','A','A','A','A','A','A','A','A','A','B','B','B','B','B'],
      'SubProd' : ['BL','BL','BL','BL','BL','lk','lk','lk','lk','lk','lk','po','po','po','po','po']}
df = pd.DataFrame(data=datas, index=Index)
print(df)

输出:

           Product SubProd
4/20/2012        A      BL
4/27/2012        A      BL
5/4/2012         A      BL
5/11/2012        A      BL
5/18/2012        A      BL
4/20/2012        A      lk
4/27/2012        A      lk
5/4/2012         A      lk
5/11/2012        A      lk
5/18/2012        A      lk
5/25/2012        A      lk
10/31/2014       B      po
11/7/2014        B      po
11/14/2014       B      po
11/21/2014       B      po
11/28/2014       B      po

谢谢

Answer 1

In [10]: df['counter'] = df.groupby(['Product','SubProd']).cumcount()+1

In [11]: df
Out[11]: 
           Product SubProd  counter
4/20/2012        A      BL        1
4/27/2012        A      BL        2
5/4/2012         A      BL        3
5/11/2012        A      BL        4
5/18/2012        A      BL        5
4/20/2012        A      lk        1
4/27/2012        A      lk        2
5/4/2012         A      lk        3
5/11/2012        A      lk        4
5/18/2012        A      lk        5
5/25/2012        A      lk        6
10/31/2014       B      po        1
11/7/2014        B      po        2
11/14/2014       B      po        3
11/21/2014       B      po        4
11/28/2014       B      po        5