熊猫：使用两列和两个数据框进行垂直查找

Question

Given the following dataframes:

df1:

    ID  A           B           
0   0   138.610513  34.860445   
2   2   139.307536  34.919052       

df2:

    ID  A           B           CAT 
0   0   138.610513  34.860445   a
1   1   138.523152  34.807862   b
2   2   139.307536  34.919052   c
3   3   138.620263  34.883671   b

How can I look up the values in CAT and add them to df1 as a new column?

I have tried this:

df1['CAT']=df1[['A'],['B']].map(df2[['A'],['B']])

But I get:

TypeError: unhashable type: 'list'

Expected output:

df1:

    ID  A           B           CAT 
0   0   138.610513  34.860445   a
2   2   139.307536  34.919052   c 

This is just a test case. In my real problem, I cannot use the IDs as a reference because they are not consistent.

Answer 1

Use merge . Looks like id too is mapped.

In [4820]: df1.merge(df2)
Out[4820]:
   ID           A          B CAT
0   0  138.610513  34.860445   a
1   2  139.307536  34.919052   c

If not, specify keys in on , and selectively pick needed columns in df2

In [4825]: df1.merge(df2[['A', 'B', 'CAT']], on=['A', 'B'])
Out[4825]:
   ID           A          B CAT
0   0  138.610513  34.860445   a
1   2  139.307536  34.919052   c