如何将数据框的列分组以在熊猫中列出？

Question

member_srl click_day productid 
0  6963 20170106 3927352 
1  6963 20170106 3790726 
2  6963 20170106  977962 
3  6963 20170106 1393860 
4  6963 20170106 3759353 

这是我的df，我想将member_srl和click_day分组，以获得productid的列表。例如，member_srl 6963和click_day 20170106将对应产品列表：[3927352,3790726,977962,1393860,3759353]

谢谢。

Answer 1

使用groupby与applylist：

df = df.groupby(['member_srl','click_day'])['productid'].apply(list) 
print (df) 
member_srl click_day 
6963  20170106  [3927352, 3790726, 977962, 1393860, 3759353] 
Name: productid, dtype: object 

df = df.groupby(['member_srl','click_day'])['productid'].apply(list).reset_index() 
print (df) 
    member_srl click_day          productid 
0  6963 20170106 [3927352, 3790726, 977962, 1393860, 3759353]