How to Slice a 2D Array into Smaller 2D Arrays with NumPy?

Slicing a 2D Array into Smaller 2D Arrays with Numpy

Often, it becomes necessary to split a 2D array into smaller 2D arrays. For instance, consider the task of dividing a 2x4 array into two 2x2 arrays.

Solution:

A combination of reshape and swapaxes functions proves effective in this scenario:

import numpy as np

def blockshaped(arr, nrows, ncols):
    """
    Converts a 2D array into a 3D array with smaller subblocks.
    """
    h, w = arr.shape
    assert h % nrows == 0, f"{h} rows is not divisible by {nrows}"
    assert w % ncols == 0, f"{w} columns is not divisible by {ncols}"

    return (arr.reshape(h//nrows, nrows, -1, ncols)
               .swapaxes(1,2)
               .reshape(-1, nrows, ncols))

Example Usage:

Consider the following array:

c = np.arange(24).reshape((4, 6))
print(c)

Output:

[[ 0  1  2  3  4  5]
 [ 6  7  8  9 10 11]
 [12 13 14 15 16 17]
 [18 19 20 21 22 23]]

Slicing this array into smaller blocks:

print(blockshaped(c, 2, 3))

Output:

[[[ 0  1  2]
  [ 6  7  8]]

 [[ 3  4  5]
  [ 9 10 11]]

 [[12 13 14]
  [18 19 20]]

 [[15 16 17]
  [21 22 23]]]

Additional Notes:

• The unblockshaped function can be used to reverse the slicing.
• Superbatfish's blockwise_view provides an alternative with its distinct format.

The above is the detailed content of How to Slice a 2D Array into Smaller 2D Arrays with NumPy?. For more information, please follow other related articles on the PHP Chinese website!