How can I combine the part-duplicate data in a DataFrame?

Question

I'm processing my data from MIMIC dataset. Some of my data are like that: (the data type is pandas.dataframe)

time  A  B  C D
01:00 2 NaN 3 4
02:00 2 NaN 3 4
03:00 2 NaN 3 4
01:00 NaN 4 3 4

NaN means missing data.

Obviously line 1 and line 4(they are token in same time) should be combined. But how can I do that?

The output I need is like:

time  A  B  C D
01:00 2  4  3 4
02:00 2 NaN 3 4
03:00 2 NaN 3 4

r.burak · Accepted Answer · 2021-02-04T10:35:32.067

0

If you want to sum the other rows , here is the code :

df.groupby(['time']).sum()

or

df.groupby(['time']).max()

edited Feb 04 '21 at 10:35

answered Feb 04 '21 at 10:32

r.burak

yep I know the sum method. But here I need to keep the other data in a row (they are same except NaN), not to sum them. – 潘嶓 Bo Pan Feb 04 '21 at 10:34
can you edit the question and show me the output you need ? – r.burak Feb 04 '21 at 10:34
ok I just added the output I need ! – 潘嶓 Bo Pan Feb 04 '21 at 10:38
ok. just use max() – r.burak Feb 04 '21 at 13:09

1 Answers1