Input is as follows:
userid,sessionid,phone,startdt,endt
1,1,samsung,03/04/22,03/05/22
1,2,iphone,03/05/22,03/06/22
1,3,htc,03/06/22,03/06/22
1,4,iphone,03/06/22,03/07/22
2,1,samsung,03/04/22,03/05/22
3,1,iphone,03/04/22,03/06/22
Output is as follows:
userid,sessionid,phone,startdt,endt
1,4,iphone,03/06/22,03/07/22
2,1,samsung,03/04/22,03/05/22
3,1,iphone,03/04/22,03/06/22
Group by user id and take max of sessionid and print the corresponding column values
The following is what I came up with
df1.groupby("userid").agg({'sessionid':'max','phone':'first'})