How sum columns with same name

Asked Jul 23 '18 at 16:22

Active Jul 23 '18 at 16:22

Viewed 2,401 times

I have a dataframe with 8000 columns and 37605 rows, I want to merge the same column names and sum the values, but as you can see the columns have a subindex(I don't know the name), the difference between X98598.1 and X98598. that I think cause me a problem. This is a part of the df

X98595      X98596      X98597      X98597.1    X98598      X98598.1
-0.460291   1.112228    -0.797426   -1.223580   -0.347824  -0.126443     
 0.233851   0.001305    -0.572343   -0.312111   -0.192996  -0.643598     
 0.003458  -1.068415    1.284535    -0.924175   -0.379626  -0.021685
 1.904167  -0.400165    -0.440025   -0.638341   -0.637231  -0.347531

and I've tried with this code

df.groupby(df.columns, axis=1).sum()

but doesn't work, do you have any idea? Please

asked Jul 23 '18 at 16:22

1

`df.T.groupby([i.split('.')[0] for i in df.T.index.values]).sum().T` – user3483203 Jul 23 '18 at 16:26
switch your `groupby` to be on `df.columns.str.split('.').str[0]` – root Jul 23 '18 at 16:28
I tried the code and it worked, why the suggestion then? – Yuca Jul 23 '18 at 16:41
It doesn't work, I still have the 8000 columns, not sum same column names – Jul 23 '18 at 17:47

How sum columns with same name

0 Answers0