Python / Pandas: How to merge rows in dataframe

Question

After merging of two data frames:

output = pd.merge(df1, df2, on='ID', how='outer')

I have data frame like this:

index  x    y   z
  0    2   NaN  3
  0   NaN   3   3
  1    2   NaN  4
  1   NaN   3   4
...

How to merge rows with the same index? Expected output:

index  x   y  z
  0    2   3  3
  1    2   3  4

What happens if the values in z differ? Does that ever happen? — IanS
– IanS, Commented Jul 18, 2017 at 9:52
You're merging on 'ID', but it isn't anywhere in your dataframe. Feels like we're missing some data. — elPastor
– elPastor, Commented Jul 18, 2017 at 9:53
This is a very simple question (in terms of reproducibility), if you want the best answers, always post a reproducible question. Include original dataframes as well. — elPastor
– elPastor, Commented Jul 18, 2017 at 10:03

Zero · Accepted Answer · 2017-07-18 09:53:54Z

3

Perhaps, you could take mean on them.

In [418]: output.groupby('index', as_index=False).mean()
Out[418]:
   index    x    y  z
0      0  2.0  3.0  3
1      1  2.0  3.0  4

answered Jul 18, 2017 at 9:53

Zero

77.4k22 gold badges153 silver badges153 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

Sevanteri · Accepted Answer · 2017-07-18 09:59:04Z

2

We can group the DataFrame by the 'index' and then... we can just get the first values with .first() or minimum with .min() etc. depending on the case of course. What do you want to get if the values in z differ?

In [28]: gr = df.groupby('index', as_index=False)

In [29]: gr.first()
Out[29]:
   index    x    y  z
0      0  2.0  3.0  3
1      1  2.0  3.0  4

In [30]: gr.max()
Out[30]:
   index    x    y  z
0      0  2.0  3.0  3
1      1  2.0  3.0  4

In [31]: gr.min()
Out[31]:
   index    x    y  z
0      0  2.0  3.0  3
1      1  2.0  3.0  4

In [32]: gr.mean()
Out[32]:
   index    x    y  z
0      0  2.0  3.0  3
1      1  2.0  3.0  4

answered Jul 18, 2017 at 9:59

Sevanteri

4,0981 gold badge26 silver badges30 bronze badges

Collectives™ on Stack Overflow

Python / Pandas: How to merge rows in dataframe

2 Answers 2

Comments

Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related