Numpy way of substracting each element in a larger matrix by a smaller matrix

Question

Is there a fast numpy way of quickly doing the following:

x = np.array([0,1,2,3,4]) # a 
y = np.array([10,20])

# x - y = np.array([[10, 20],[9,19],[8,18],[7,17],[6,16]]) # 5x2 matrix

Where the result would be the absolute difference between each element in x minus each element in y

Just like mathematical matrices, the two arrays need to be the same shape and size. You will need to vstack one of the arrays and hstack the other. — Beefster, Dec 28 '17 at 20:34

score 0 · Accepted Answer · answered Dec 28 '17 at 21:11

0

Yes there is, its down to how you build your array containing the result...

import numpy as np

x = np.array([0,1,2,3,4]) 
y = np.array([10,20])
z = np.array([y[0]-x[:], y[1]-x[:]])
zT = np.transpose(z)

answered Dec 28 '17 at 21:11

DrBwts

actually Paul Panzer's answer in the comments under the question is a much better solution – DrBwts Dec 28 '17 at 21:14

1 Answers1