Rearranging 2D numpy array by 2D row and column arrays

Question

I have tried to find a similar question but so far it seems only half my question can be answered.

I have a 2D numpy array, e.g.:

a= np.array([[6, 4, 5],
             [4, 7, 8],
             [2, 8, 9]])

And i also have 2 further numpy arrays, indicating the rows, and columns where i would like to rearrange (or not):

rows= np.array([[0, 0, 0],
                [1, 0, 1],
                [2, 2, 2]])

cols= np.array([[0, 1, 2],
                [0, 0, 2],
                [0, 1, 2]])

now i would like to rearrange the array "a" based on these indices, so that the result is:

result= np.array([[6, 4, 5],
                  [4, 6, 8],
                  [2, 8, 9]])

Doing this only for columns or only for rows is easy, e.g. see this Thread:

np.array(list(map(lambda x, y: y[x], cols, a)))

Believe it or not, you can simply write `a[rows, cols]` :) – Chrysophylaxs Feb 14 '23 at 10:53 — Chrysophylaxs, Feb 14 '23 at 10:53

score 2 · Answer 1 · answered Feb 14 '23 at 11:42

2

This is a typical case of fancy/array indexing:

result = a[rows, cols]

Output:

array([[6, 4, 5],
       [4, 6, 8],
       [2, 8, 9]])

answered Feb 14 '23 at 11:42

mozway

194,879
13
39
75

1

so simple- so great... – shmulik90 Feb 14 '23 at 12:30
@SamuelMüller indeed :) – mozway Feb 14 '23 at 13:05

Rearranging 2D numpy array by 2D row and column arrays

1 Answers1