Reformat output of numpy.where

Asked Jun 15 '17 at 20:53

Active Jun 15 '17 at 20:56

Viewed 36 times

The following code

arr_0 = numpy.zeros((3,3))
arr_0[(0,0)] = 1
ind = numpy.where(arr_0 == 0)
print(ind)

gives the following output

(array([0, 0, 1, 1, 1, 2, 2, 2], dtype=int64), array([1, 2, 0, 1, 2, 0, 1, 2], dtype=int64))

The output is a list of x's and then a list of y's, and I would like a list of tuples (x,y), something like: [[0,1],[0,2], ...]. How can I do that ?

edited Jun 15 '17 at 20:56

asked Jun 15 '17 at 20:53

arnold_107

A list of tuples? Or a numpy matrix? – Willem Van Onsem Jun 15 '17 at 20:54
Maybe, the idea is to be able to select a tuple (x,y) randomly from the output. – arnold_107 Jun 15 '17 at 20:55
2

Use `list(zip(*ind))`. – ayhan Jun 15 '17 at 20:56
Ok, so the term is "merge lists". – arnold_107 Jun 15 '17 at 20:58
Yes, that seems like the most common term for this (among pair up, combine, paste together, pairwise append, etc.). :) https://stackoverflow.com/questions/linked/2407398?page=1&sort=hot – ayhan Jun 15 '17 at 21:08
3

Try `np.transpose(ind)`. Or `np.argwhere(...)`, which does the same thing. – hpaulj Jun 15 '17 at 21:18

0 Answers0