How do I merge several 2D arrays into a single 2D array (list) in a box-like manner in python?

Say I have 9 2D arrays in the following format:

A1 = [[ a1, b1, c1 ],
      [ d1, e1, f1 ],
      [ g1, h1, i1 ]]

A2 = [[ a2, b2, c2 ],
      [ d2, e2, f2 ],
      [ g2, h2, i2 ]] 
.....
A9 = [[ a9, b9, c9 ],
      [ d9, e9, f9 ],
      [ g9, h9, i9 ]]

and I want to concatenate them to get a 2D array like this:

A = [B1, B2, B3]

where

B1 = np.concatenate((A1,A2, A3),axis=1) 
B2 = np.concatenate((A4,A5, A6),axis=1) 
B3 = np.concatenate((A7,A8, A9),axis=1) 

I will have N arrays in my case and I calculate the value of N like so:

img = Image.open(file_name)
img_width, img_height = img.size

tile_height = int(input('Enter the height of tile:'))
tile_width = int(input("Enter the width of tile:'))

N = (img_height//tile_height)*(img_width//tile_width)

# **The image will be broken down into n tiles of size tile_width x tile_height**

for i in range(img_height//tile_height):
    for j in range(img_width//tile_width):
         box = (j*width, i*height, (j+1)*width, (i+1)*height)
         img.crop(box)
         ...

So essentially, I had an image that was broken down into N tiles and after some processing, I have these image tile data stored as numpy arrays and I want to concatenate/merge them into a single 2D numpy array in the same orientation as the original image. How can I do that?

Answer

This seems a perfect use case for bmat

Edit: How to use bmat

bmat accepts as the first argument a matrix of blocks.

[[A11, A12, ..., A1n]
 [A21, A22, ..., A2n]
 ...
 [Am1, Am2, ..., Amn]]

And is not limited to the 9 submatrices case, it was a coincidence that in the bmat documentation their example is the same size as the example in your question.

import numpy as np;
mats = []
for i in range(10):
    mats.append(np.ones((4, 2)) * i);
np.bmat([mats[:5], mats[5:]])

Gives

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