Efficiently test matrix rows and columns with numpy

Question

I am trying to remove both the row i and column i when both the row i and column i contains all 0s. For example in this case we can see that row 0 is all zeros

Answer 1

One thing you can do- evaluate the boolean matrix "matrix == 0" outside of your for loop, and use that matrix instead of reevaluating it every run.

It'd be something like:

def remove(matrix):
    binary_matrix = matrix == 0
    for i, x in reversed(list(enumerate(matrix))):
        if np.all(binary_matrix , axis=0)[i] and np.all(binary_matrix , axis=1)[i]:
            matrix = np.delete(matrix,i,axis=0)
            matrix = np.delete(matrix,i,axis=1)
            binary_matrix = np.delete(binary_matrix ,i,axis=0)
            binary_matrix = np.delete(binary_matrix ,i,axis=1)
    return matrix