GROUPBY - 蟒大熊猫数据帧

Question

我有列date，name，id（数据是冗余的）一个数据帧。
现在我想获得组合的频率（date，name，id） 对于我在我的数据框应用groupby：

df.groupby(['date','uname','id']).size() 

这是给我造成这样的：

date  uname    id  size       
2016-02-11 [email protected]   111  1 
      [email protected]   1080  2 
      [email protected]   5315  1 
           5317  1 
      [email protected]  536  2 
2-16-02-12 [email protected]  588  1 
           593  2  
      [email protected]  322  1 

但我想我的结果是这样的：

date  uname    id  size       
2016-02-11 [email protected]   111  1 
2016-02-11 [email protected]   1080  2 
2016-02-11 [email protected]   5315  1 
2016-02-11 [email protected]   5317  1 
2016-02-11 [email protected]  536  2 
2-16-02-12 [email protected]  588  1 
2-16-02-12 [email protected]  593  2  
2-16-02-12 [email protected]  322  1 

欣赏建议

Answer 1

您需要reset_index：

df.groupby(['date','uname','id']).size().reset_index() 

     date    uname id size 
0 2016-02-11   [email protected] 111  1 
1 2016-02-11   [email protected] 1080  2 
2 2016-02-11  [email protected] 5315  1 
3 2016-02-11  [email protected] 5317  1 
4 2016-02-11 [email protected] 536  2 
5 2-16-02-12 [email protected] 588  1 
6 2-16-02-12 [email protected] 593  2 
7 2-16-02-12 [email protected] 322  1