how to use python-numpy-broadcasting

Question

I have a trouble with python broadcasting

there is two numpy list

x = np.array([[1,2,3],[4,5,6]])

y = np.array([0,1,1])

I'd like to calculate

x : [[1,2,3],
     [4,5,6]]

to

x : [[1-y[0], 2-y[1], 3-y[2]],
     [4-y[0], 5-y[1], 6-y[2]]]

that is,

x : [[0,-3,-3],
     [3,0,0]]

at one time.

Please, let me know how to do that.

I'm sorry...i had a mistake....

i mean x : [[1,1,2], [4,4,5]]

check your math dude! a usual prerequisite for programming is to be able to do integer subtraction... — Julien, Jul 07 '16 at 05:42
[[1-0, 2-1, 3-1],[4-0, 5-1, 6-1]] == [[0, -3, -3], [3, 0, 0]] ??? — Julien, Jul 07 '16 at 05:44
Thanks for your comments guys. But, I want to calculate with no loop. — yi han, Jul 08 '16 at 06:54

Gal Dreiman · Answer 1 · 2016-07-07T07:08:13.137

2

I don't see how you got:

x : [[0,-3,-3],
     [3,0,0]]

But the easiest way to do what you ask for is just:

import numpy as np
X = ([1,2,3],[4,5,6])
y = np.array([0,1,1])
print X-y

Output:

[ [1 1 2]
[4 4 5] ]

edited Jul 07 '16 at 07:08

answered Jul 07 '16 at 06:28

Gal Dreiman

3

Why are you iterating? Just do `X-y` and get a 2d array. Since `y` is an array, it treats `X` as one as well. – hpaulj Jul 07 '16 at 06:33
Thanks. it's very easy. – yi han Jul 08 '16 at 07:05

1 Answers1