Applying .isin for multiple series in a DF

Question

I have three fields TEL1, TEL2, TEL3 which I want to look to see if values exist in another DF. I can do one column at a time, but was wondering if there was a quick way I could hide rows if any of the three fields contains a value in the lookup Dataframe.

I am currently using:

df1 = Sample['TEL1'].isin(DNC['NUMBER'])

Dataframe I am looking up against:

 NUMBER 0 555

Dataframe I am looking up:

 TEL1 TEL2 TEL3 NAME 0 222 555 555 BEN 1 222 555 222 BEN 2 555 222 555 BEN 3 222 222 222 BEN 4 555 555 555 BEN

Expected result (row that does not contain 555 in TEL1-TEL3 columns):

 TEL1 TEL2 TEL3 NAME 3 222 222 222 BEN

Thank you in advance, Ben.

busybear · Accepted Answer · 2019-02-16 03:51:18Z

You can us isin on the entire dataframe (or with the columns you are interested in) and then check which rows are all False (ie, not have any).

ix = ~df.loc[:, 'TEL1':'TEL3'].isin(DNC['NUMBER'].values).any(axis=1) df.loc[ix]

Notice I used the values of DNC['NUMBER'] because it will try to match the index otherwise. So it'll look for 555 in the dataframe where index is 0 if you keep it as a series.

Thanks so much this is perfect. The numbers will change from just 555, so it's great that it can still be dynamic.

Collectives™ on Stack Overflow

Applying .isin for multiple series in a DF

1 Answer 1

1 Comment

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

1 Comment

Related