counting surrounding true's in numpy matrix (python)

Question

I want to make something similiar to minesweeper

input matrix:

matrix = [[true, false, false],
          [false, true, false],
          [false, false, false]]

if a bomb is on the field I do not count it as bomb is surrounding.

I thought about doing it with numpy convolve somehow but I am struggling with how to go through the matrix and always check the left, top, right and bottom fields of the actual field (in the case of a border I check "empty" fields which is 0 for sure)

no it doesn't have to be numpy, but I thought that this would be the best solution, just not sure how to get it to work. @DanielF hmm how that? — Dominik Lemberger, Aug 10 '18 at 08:58

ibarrond · Accepted Answer · 2018-08-10T09:52:08.483

5

Here is a solution using scipy.signal.convolve2d :

import scipy
import numpy as np

# Input matrix, can be left as boolean
matrix = np.array([[True, False, False],
                   [False, True, False],
                   [False, False, False]])

# Our dougnut filter
W = np.array([[1, 1, 1],
              [1, 0, 1],
              [1, 1, 1]])

# Single convolve
res = convolve2d(matrix, W, 'same')

We get the exact result:

res
array([[1, 2, 1],
       [2, 1, 1],
       [1, 1, 1]])

edited Aug 10 '18 at 09:52

answered Aug 10 '18 at 09:03

ibarrond

6,617
4
26
45

Dang, got FGITW'd. You can do all the padding and slicing using the `boundary` and `mode` parameters of `convolve2d` though. – Daniel F Aug 10 '18 at 09:14
mode=same is enough, doesnt need the boundary tho :P - boundary wrap makes the output actually wrong as it makes 1,2,1 to 2,1,2 – Dominik Lemberger Aug 10 '18 at 09:24
Updated! no more boundary – ibarrond Aug 10 '18 at 09:52

score 1 · Answer 2 · answered Aug 10 '18 at 09:07

using only numpy, including nd_window from here

m_pad = np.pad(matrix, ((1,1),(1,1)), 'constant', constant_values=(False, False))
filter = np.array([[1,1,1],[1,0,1],[1,1,1]], dtype = bool)
adj_matrix = np.einsum('ijkl,kl->ij', nd_window(m_pad, 3), filter)

Also, you can use scipy.signal.convolve2d

adj_matrix = convolve2d(matrix, filter, 'same', fillvalue = False)

score 0 · Answer 3 · answered Aug 10 '18 at 09:12

To keep it simple : using numpy you can simply use the sum() method (https://docs.scipy.org/doc/numpy-1.13.0/reference/generated/numpy.matrix.sum.html).

np.matrix(matrix).sum() will count all the True in your matrix.

So np.matrix(matrix).sum() - matrix[1][1] should return the number of mine around a cell.

And for me the easiest way to remove the border problem is to complete your matrix with False cells all around it.

counting surrounding true's in numpy matrix (python)

3 Answers3