CODEWITHSUNDEEP
pythonarraysnumpysum
Fred vh
Fred vh

Reputation: 47

Forming a matrix by summing 2 arrays element-array wise using NumPy

Let's say I have the following NumPy arrays:

i = array([2, 4, 5])
j = array([0, 1, 2])

I would like to have a very efficient method (built-in if possible) to sum those vectors and have an output that looks like this:

[[2 4 5]
 [3 5 6]
 [4 6 7]]

So basically each column is the array j to which the k th element of i has been added (k = 0, 1, 2 in this case)

Upvotes: 2

Views: 70

Answers (2)

hpaulj
hpaulj

Reputation: 231738

Or with broadcasting:

In [272]: i[:,None]+j
Out[272]: 
array([[2, 3, 4],
       [4, 5, 6],
       [5, 6, 7]])

i[:,None] makes a (3,1) array, which broadcasts with a (3,) (or (1,3)) to make a (3,3).

Upvotes: 3

timgeb
timgeb

Reputation: 78800

Use numpy.add.outer.

>>> import numpy as np                                                                                                 
>>> i = np.array([2, 4, 5])                                                                                            
>>> j = np.array([0, 1, 2])                                                                                            
>>>                                                                                                                    
>>> np.add.outer(j, i)                                                                                                 
array([[2, 4, 5],
       [3, 5, 6],
       [4, 6, 7]])

Upvotes: 3

Related Questions