How to move items in a numpy array in Python?

Question

I haven't found a simple solution to move elements in a NumPy array.

Given an array, for example:

>>> A = np.arange(10).reshape(2,5)
>>> A
array([[0, 1, 2, 3, 4],
       [5, 6, 7, 8, 9]])

and given the indexes of the elements (columns in this case) to move, for example [2,4], I want to move them to a certain position and the consecutive places, for example to p = 1, shifting the other elements to the right. The result should be the following:

array([[0, 2, 4, 1, 3],
       [5, 7, 9, 6, 8]])

yes, but then it is the same problem with the indexes. How can I do this automatically? — Diego Palacios
– Diego Palacios, Commented Feb 3, 2021 at 11:02

Stef · Accepted Answer · 2021-02-03 17:57:28Z

1

You can create a mask m for the sorting order. First we set the columns < p to -1, then the to be inserted columns to 0, the remaining columns remain at 1. The default sorting kind 'quicksort' is not stable, so to be safe we specify kind='stable' when using argsort to sort the mask and create a new array from that mask:

import numpy as np

A = np.arange(10).reshape(2,5)
p = 1
c = [2,4]

m = np.full(A.shape[1], 1)
m[:p] = -1    # leave up to position p as is
m[c] = 0      # insert columns c

print(A[:,m.argsort(kind='stable')])
#[[0 2 4 1 3]
# [5 7 9 6 8]]

edited Feb 3, 2021 at 17:57

answered Feb 3, 2021 at 12:04

Stef

30.9k3 gold badges34 silver badges60 bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

Diego Palacios Over a year ago

I think quicksort is not stable.

Stef Over a year ago

@DiegoPalacios yes, my bad. I updated the answer.

Collectives™ on Stack Overflow

How to move items in a numpy array in Python?

1 Answer 1

2 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

2 Comments

Your Answer

Sign up or log in

Post as a guest

Related