CODEWITHSUNDEEP
pythonpandas
David Yang
David Yang

Reputation: 2141

Pandas : Merge two Dataframes on a column, but keep only data for unlike Columns

I have 2 DataFrames: frame1, and frame2

In [10]: frame1[:5]
Out[10]:
    cid
0   531
1  1102
2  1103
3  1406
4  1409

In [14]: frame2[:5]
Out[14]:
     cid   media_cost     imps  booked_revenue
0  72692    29.671446    13918       84.961853
1  72704  3121.781201  6992946     9912.982516
2    531     0.001540        2        0.000000
3  39964  2307.119001  3997167     5425.629736
4  72736    45.716847   143574       56.280000

frame1 has 60,888 rows, frame2 has 139,846 rows.

Using these two Dataframes, I want to create a third Dataframe that basically consists of frame2 with all the cid values it shares with frame1 removed. So, in this example, I would want a frame3 that is frame2 without row 2, cid 531, that it shares with frame1.

Upvotes: 1

Views: 766

Answers (1)

DSM
DSM

Reputation: 353179

How about:

>>> f1
    cid
0   531
1  1102
2  1103
3  1406
4  1409
>>> f2
     cid   media_cost     imps  booked_revenue
0  72692    29.671446    13918       84.961853
1  72704  3121.781201  6992946     9912.982516
2    531     0.001540        2        0.000000
3  39964  2307.119001  3997167     5425.629736
4  72736    45.716847   143574       56.280000
>>> f2[~f2.cid.isin(f1.cid)]
     cid   media_cost     imps  booked_revenue
0  72692    29.671446    13918       84.961853
1  72704  3121.781201  6992946     9912.982516
3  39964  2307.119001  3997167     5425.629736
4  72736    45.716847   143574       56.280000

Upvotes: 3

Related Questions