1

I have 9x9 matrix in numpy. I need fast method to change it into sudoku-like 9x9 array with cubes (also I need method to reverse this operation). I attach conceptual drawing.

conceptual drawing

minami1120
  • 11
  • 1
  • 1
    Welcome to StackOverflow! That's an interesting question. What did you try and/or research? Why didn't that work for you? – Jondiedoop Mar 24 '19 at 08:30
  • Hi! The original form of matrix is easier for me to work with (doing checks etc.) but I need sudoku-like representation for convolutional neural network – minami1120 Mar 24 '19 at 08:49
  • Did the posted solution work for you? – Divakar Mar 24 '19 at 13:57

1 Answers1

0

Reshape, permute axes and reshape -

In [43]: a
Out[43]: 
array([[0, 1, 2, 3, 4, 5, 6, 7, 8],
       [0, 1, 2, 3, 4, 5, 6, 7, 8],
       [0, 1, 2, 3, 4, 5, 6, 7, 8],
       [0, 1, 2, 3, 4, 5, 6, 7, 8],
       [0, 1, 2, 3, 4, 5, 6, 7, 8],
       [0, 1, 2, 3, 4, 5, 6, 7, 8],
       [0, 1, 2, 3, 4, 5, 6, 7, 8],
       [0, 1, 2, 3, 4, 5, 6, 7, 8],
       [0, 1, 2, 3, 4, 5, 6, 7, 8]])

In [44]: a.reshape(3,3,3,3).swapaxes(1,2).reshape(9,9)
Out[44]: 
array([[0, 1, 2, 0, 1, 2, 0, 1, 2],
       [3, 4, 5, 3, 4, 5, 3, 4, 5],
       [6, 7, 8, 6, 7, 8, 6, 7, 8],
       [0, 1, 2, 0, 1, 2, 0, 1, 2],
       [3, 4, 5, 3, 4, 5, 3, 4, 5],
       [6, 7, 8, 6, 7, 8, 6, 7, 8],
       [0, 1, 2, 0, 1, 2, 0, 1, 2],
       [3, 4, 5, 3, 4, 5, 3, 4, 5],
       [6, 7, 8, 6, 7, 8, 6, 7, 8]])

On a generic shaped array, it would be -

m,n = a.shape
H,W = (3,3) # block size on sudoku
out = a.reshape(m//H,H,n//W,W).swapaxes(1,2).reshape(m,n)

More info on intuition behind nd-to-nd array transformation.

If you want to start from scratch, use kron on a ranged-array -

In [65]: np.kron(np.ones((3,3),dtype=int),np.arange(9).reshape(3,3))
Out[65]: 
array([[0, 1, 2, 0, 1, 2, 0, 1, 2],
       [3, 4, 5, 3, 4, 5, 3, 4, 5],
       [6, 7, 8, 6, 7, 8, 6, 7, 8],
       [0, 1, 2, 0, 1, 2, 0, 1, 2],
       [3, 4, 5, 3, 4, 5, 3, 4, 5],
       [6, 7, 8, 6, 7, 8, 6, 7, 8],
       [0, 1, 2, 0, 1, 2, 0, 1, 2],
       [3, 4, 5, 3, 4, 5, 3, 4, 5],
       [6, 7, 8, 6, 7, 8, 6, 7, 8]])
Divakar
  • 218,885
  • 19
  • 262
  • 358