Numpy summation of 2D array [duplicate]

Question

I have two relatively similar operations I'd like to figure out how to do.

The first is the simpler. I have a numpy array like

[[a,b,c,d],
 [e,f,g,h],
 [i,j,k,l]]

I would like to sum the columns to yield:

[6,12,18,24]

I know that I can do this with a for loop:

for i in range(0,ROW_LENGTH): #ROW_LENGTH defined elsewhere
    listsum[i]=list[:,i].sum()

But this can't be the pythonic way, can it?

The second task is I need to do something similar with a 3D array. I assume the solution to this will be similar. A simple example is:

[[[a,b,c],[d,e,f],[g,h,i]],
 [[j,k,l],[m,n,o],[p,q,r]],
 [[s,t,u],[v,w,x],[y,z,_]]]

Where the output should be:

[[a+j+s,b+k+t,c+l+u],[d+m+v,e+n+w,f+o+x],[g+p+y,h+q+z,i+r+_]]

I'm hopeful that understanding the technique for both of these will be a major boon, generally speaking, for how well I grasp python!

I'm going to link here to a separate question I'm asking for the same project:

Do you know about the axis parameter in np.sum? It allows you to perform these operation very efficiently. — Alex Riley
– Alex Riley, Commented Oct 28, 2015 at 15:38
WOW. Yes. Exactly. So what I want is np.sum(list,axis=0), correct? Can you put this in answer form so I can upvote and select? This also solves the other problem I was going to link to, which was with a product instead. THANK YOU! — Frank Harris
– Frank Harris, Commented Oct 28, 2015 at 15:53
Glad it helped - I've marked this question as a duplicate so it can help direct users to the correct answer. — Alex Riley
– Alex Riley, Commented Oct 28, 2015 at 16:03

dan_nodescu · Accepted Answer · 2015-10-28 16:00:41Z

0

In order to do that you can use the sum method of numpy.

import numpy as np
a = np.array([[1,2,3],[4,5,6],[7,8,9]])
a.sum(axis=0)

That should do the trick

answered Oct 28, 2015 at 16:00

dan_nodescu

1

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

1 Answer 1