问题描述
这是我第一次在熊猫中使用多重索引，我需要一些帮助来合并两个带有层次结构列的数据框。 这是我的两个数据框：
col_index = pd.MultiIndex.from_product([['a','b','c'],['w','x']])
df1 = pd.DataFrame(np.ones([4,6]),columns=col_index,index=range(4))
a b c
w x w x w x
0 1.0 1.0 1.0 1.0 1.0 1.0
1 1.0 1.0 1.0 1.0 1.0 1.0
2 1.0 1.0 1.0 1.0 1.0 1.0
3 1.0 1.0 1.0 1.0 1.0 1.0
df2 = pd.DataFrame(np.zeros([2,index=range(2))
a b c
w x w x w x
0 0.0 0.0 0.0 0.0 0.0 0.0
1 0.0 0.0 0.0 0.0 0.0 0.0
使用merge方法时，会得到以下结果：
pd.merge(df1,df2,how='left',suffixes=('','_2'),left_index = True,right_index= True ))
a b c a_2 b_2 c_2
w x w x w x w x w x w x
0 1.0 1.0 1.0 1.0 1.0 1.0 0.0 0.0 0.0 0.0 0.0 0.0
1 1.0 1.0 1.0 1.0 1.0 1.0 0.0 0.0 0.0 0.0 0.0 0.0
2 1.0 1.0 1.0 1.0 1.0 1.0 NaN NaN NaN NaN NaN NaN
3 1.0 1.0 1.0 1.0 1.0 1.0 NaN NaN NaN NaN NaN NaN
但是我想合并两个较低级别的数据框，使后缀对['w'，'x'] 生效，如下所示：
a b c
w w_2 x x_2 w w_2 x x_2 w w_2 x x_2
0 1.0 0.0 1.0 0.0 1.0 0.0 1.0 0.0 1.0 0.0 1.0 0.0
1 1.0 0.0 1.0 0.0 1.0 0.0 1.0 0.0 1.0 0.0 1.0 0.0
2 1.0 NaN 1.0 NaN 1.0 NaN 1.0 NaN 1.0 NaN 1.0 NaN
3 1.0 NaN 1.0 NaN 1.0 NaN 1.0 NaN 1.0 NaN 1.0 NaN
解决方法
您可以将join或merge与swaplevel()或reorder_levels一起使用。然后使用.sort_index()并传递axis=1来按索引列排序。
-
当您对这样的索引进行合并时，
-
.join()会更好。 -
.swaplevel()在有两个级别时（在这种情况下）更好，而.reorder_levels()在三个或三个以上级别时更好。
以下是这些方法的4种组合。对于这个特定的示例，我认为.join() / .swaplevel()是最容易出现的情况（请参见最后一个示例）：
df3 = (df1.reorder_levels([1,0],axis=1)
.join(df2.reorder_levels([1,axis=1),rsuffix='_2')
.reorder_levels([1,axis=1).sort_index(axis=1,level=[0,1]))
df3
Out[1]:
a b c
w w_2 x x_2 w w_2 x x_2 w w_2 x x_2
0 1.0 0.0 1.0 0.0 1.0 0.0 1.0 0.0 1.0 0.0 1.0 0.0
1 1.0 0.0 1.0 0.0 1.0 0.0 1.0 0.0 1.0 0.0 1.0 0.0
2 1.0 NaN 1.0 NaN 1.0 NaN 1.0 NaN 1.0 NaN 1.0 NaN
3 1.0 NaN 1.0 NaN 1.0 NaN 1.0 NaN 1.0 NaN 1.0 NaN
df3 = (pd.merge(df1.reorder_levels([1,df2.reorder_levels([1,how='left',left_index=True,right_index=True,suffixes = ('','_2'))
.reorder_levels([1,1]))
df3
Out[2]:
a b c
w w_2 x x_2 w w_2 x x_2 w w_2 x x_2
0 1.0 0.0 1.0 0.0 1.0 0.0 1.0 0.0 1.0 0.0 1.0 0.0
1 1.0 0.0 1.0 0.0 1.0 0.0 1.0 0.0 1.0 0.0 1.0 0.0
2 1.0 NaN 1.0 NaN 1.0 NaN 1.0 NaN 1.0 NaN 1.0 NaN
3 1.0 NaN 1.0 NaN 1.0 NaN 1.0 NaN 1.0 NaN 1.0 NaN
df3 = (pd.merge(df1.swaplevel(axis=1),df2.swaplevel(axis=1),'_2'))
.swaplevel(axis=1).sort_index(axis=1,1]))
df3
Out[3]:
a b c
w w_2 x x_2 w w_2 x x_2 w w_2 x x_2
0 1.0 0.0 1.0 0.0 1.0 0.0 1.0 0.0 1.0 0.0 1.0 0.0
1 1.0 0.0 1.0 0.0 1.0 0.0 1.0 0.0 1.0 0.0 1.0 0.0
2 1.0 NaN 1.0 NaN 1.0 NaN 1.0 NaN 1.0 NaN 1.0 NaN
3 1.0 NaN 1.0 NaN 1.0 NaN 1.0 NaN 1.0 NaN 1.0 NaN
df3 = (df1.swaplevel(i=0,j=1,axis=1)
.join(df2.swaplevel(axis=1),rsuffix='_2')
.swaplevel(axis=1).sort_index(axis=1,1]))
df3
Out[4]:
a b c
w w_2 x x_2 w w_2 x x_2 w w_2 x x_2
0 1.0 0.0 1.0 0.0 1.0 0.0 1.0 0.0 1.0 0.0 1.0 0.0
1 1.0 0.0 1.0 0.0 1.0 0.0 1.0 0.0 1.0 0.0 1.0 0.0
2 1.0 NaN 1.0 NaN 1.0 NaN 1.0 NaN 1.0 NaN 1.0 NaN
3 1.0 NaN 1.0 NaN 1.0 NaN 1.0 NaN 1.0 NaN 1.0 NaN