Filtering pandas based on value tuples for multiple columns

Asked May 16 '18 at 18:34

Active May 16 '18 at 18:46

Viewed 161 times

I have a large dataframe with the following format:

user1    user2 +++++ other columns ++++++
---------------
Alice    Carol +++++ other columns ++++++
Alice    Bob   +++++ other columns ++++++
Bob      Carol +++++ other columns ++++++
Alice    Carol +++++ other columns ++++++

And I have a separate text file that contains

Alice,Bob
Bob,Carol

How do I get the output

user1    user2 +++++ other columns ++++++
---------------
Alice    Carol   +++++ other columns ++++++
Bob      Carol   +++++ other columns ++++++
Alice    Carol   +++++ other columns ++++++

without having to use df.iterrows() (since that will be slow)?

EDIT: I've also tried merge. It now works:

df = pd.DataFrame({'a':['Alice','Alice','Bob','Alice'],'b':['Carol','Bob','Carol','Carol']})
df2 = pd.DataFrame({'a':['Alice','Bob'],'b':['Carol','Carol']})
pd.merge(df,df2,on=['a','b'])

resulted in

       a      b
0  Alice  Carol
1  Alice  Carol
2    Bob  Carol

edited May 16 '18 at 18:46

asked May 16 '18 at 18:34

irene

2,085
1
22
36

1

Merge on the two columns. – cs95 May 16 '18 at 18:36
Please provide more details if merge does not work: show your code, what error did you get? – IanS May 16 '18 at 18:46
Hi @IanS edited again to show details – irene May 16 '18 at 18:47
1

Oh found my mistake. The merge should be on both 'a' and 'b'. My bad. – irene May 16 '18 at 18:48

Filtering pandas based on value tuples for multiple columns

0 Answers0