Check if all sides of a multidimensional numpy array are arrays of zeros

Question

An n-dimensional array has 2n sides (a 1-dimensional array has 2 endpoints; a 2-dimensional array has 4 sides or edges; a 3-dimensional array has 6 2-dimensional faces; a 4-dime

Answer 1

You can make use of slice and boolean masking to get the job done:

def get_borders(arr):
    s=tuple(slice(1,i-1) for i in a.shape)
    mask = np.ones(arr.shape, dtype=bool)
    mask[s] = False
    return(arr[mask])

This function first shapes the "core" of the array into the tuple s, and then builds a mask that shows True only for the bordering points. Boolean indexing then delivers the border points.

Working example:

a = np.arange(16).reshape((4,4))

print(a)
array([[ 0,  1,  2,  3],
       [ 4,  5,  6,  7],
       [ 8,  9, 10, 11],
       [12, 13, 14, 15]])

borders = get_borders(a)
print(borders)
array([ 0,  1,  2,  3,  4,  7,  8, 11, 12, 13, 14, 15])

Then, np.all(borders==0) will give you the desired information.

---

Note: this breaks for one-dimensional arrays, though I consider those an edge case. You're probably better off just checking the two points in question there