Check value of multiple columns and then assign vaue to column

Asked Oct 08 '19 at 06:57

Active Oct 08 '19 at 07:46

Viewed 20 times

This is my dataframe:

C1 | C2 |... email_id | subject | sender | recipient
   |    |             | congrats| x      |  y
   |    |             | congrats | z     | y
   |    |             | congrats | x     |y
   |    |             | meeting  | x     | y

Output:

   C1  | C2 |... email_id  | subject  | sender | recipient
       |    |       0      | congrats | x      | y
       |    |       1      | congrats | z      | y
       |    |       0      | congrats | x      | y
       |    |       2      | meeting  | x      | y

For every unique combination of subject, sender and recipient I want to assign an emai_id. I have gotten unique triples like this:

df1 = df.drop_duplicates(subset=['sender','recipient','subject'])

In order to assign value this is what I am doing

sender = df1.sender
recipient = df1.recipient
subject = df1.subject
n = 0
for i in sender:
     df.loc[df["subject"] == str(i) and df["subject"] == str(i) , "email_id"] = n
     n=n+1

This isn't the correct way to go about it. How do I add an "add" condition here? Edit: In short, assign same number to same triplets

edited Oct 08 '19 at 07:46

asked Oct 08 '19 at 06:57

Fatima Arshad

I think you need merge by left join. – jezrael Oct 08 '19 at 07:04
How would that assign value to email_id? – Fatima Arshad Oct 08 '19 at 07:05
If not working merge, please add some sample data - [minimal, complete, and verifiable example](http://stackoverflow.com/help/mcve), your merge code and expected output. – jezrael Oct 08 '19 at 07:32
@jezrael check again – Fatima Arshad Oct 08 '19 at 07:36
hmmm, so need `df['email_id'] = df.groupby(subset=['sender','recipient','subject'], sort=False).ngroup()` ? – jezrael Oct 08 '19 at 07:38
What is ngroup()? @jezrael – Fatima Arshad Oct 08 '19 at 07:42
Changed dupe answer. – jezrael Oct 08 '19 at 07:45
So change data for [minimal, complete, and verifiable example](http://stackoverflow.com/help/mcve) and explain why solution with ngroup not working. – jezrael Oct 08 '19 at 07:47
1

@jezrael it is working fine. I was printing the wrong df. Thanks – Fatima Arshad Oct 08 '19 at 07:53

Check value of multiple columns and then assign vaue to column

0 Answers0