Slicing matrix in for loop with index in python

Question

I have a 3D matrix such as:

array([[[3., 0., 0.],
        [0., 0., 0.],
        [0., 0., 0.]],
       [[0., 0., 0.],
        [2., 0., 0.],
        [0., 0., 0.]],
       [[0., 0., 0.],
        [12., 0., 0.],
        [0., 0., 0.]]])

I want to slice this in [: , 0, :], [-1,: :].. and all 6 directions in the order with a for loop. So for each dimension, slicing from the first (0) and last (-1).

What is the proper way of applying the for loop?

Lets assume the name of the array is A:

A[0, :, :]
A[:, :, 0]
A[:, 0, :]
A[-1, :, :]
A[:, -1, :]
A[:, :, -1]

I want to have these 6 submatrices (lets say in a list) in one loop.

what do you mean by "all 6 directions" and "in the order"? It might help if you give an example of the output you expect. — overfull hbox
– overfull hbox, Commented Oct 11, 2019 at 16:17
You almost never need loops with NumPy. The best answer is very unlikely to involve loops. It would help if you show what output you expect from your input? — Matt Hall
– Matt Hall, Commented Oct 11, 2019 at 16:27
This is a (3,3,3) shaped array. You need to be more explicit about what you want to do - examples of the resulting slices. Often the action is clearer if you use an array like np.arange(24).reshape(2,3,4) - distinct dimensions, and values. — hpaulj
– hpaulj, Commented Oct 11, 2019 at 19:06
edited the questions to make it a bit clearer. Reshaping is not an option. — sariug
– sariug, Commented Oct 11, 2019 at 19:30

Andras Deak -- Слава Україні · Accepted Answer · 2019-10-25 22:48:35Z

2

It's not clear at all why you want to do this, but you can dynamically snatch subarrays using take:

import numpy as np
arr = np.arange(2*3*4).reshape(2, 3, 4)  # dummy input

arrays = [arr.take(index, axis) for axis in range(arr.ndim) for index in [0,-1]]

This list of arrays will contain arr[0,:,:], arr[-1,:,:], arr[:,0,:], arr[:,-1,:] etc. in this order. Here's proof for a single element in the list of arrays:

>>> np.array_equal(arr[:,-1,:], arr.take(-1, 1))
True

Your question wasn't too specific about the order, but the above can easily be generalized for an arbitrary ordering of your slicing requirements:

indices = [0, 0, 0, -1, -1, -1]
axes = [0, 2, 1, 0, 1, 2]
arrays = [arr.take(index, axis) for index,axis in zip(indices, axes)]

This matches the order of the subarrays as specified in your question.

answered Oct 25, 2019 at 22:48

Andras Deak -- Слава Україні

35.4k13 gold badges94 silver badges118 bronze badges

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

Collectives™ on Stack Overflow

Slicing matrix in for loop with index in python

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related