Iterating over two 2D numpy arrays with index

Question

I have two 2D numpy arrays of the same shape. Is there a way to iterate through them simultaneously with getting eg a pair of elemets from both tables and their index?

For example, I have two arrays

before = np.array(
    [[0, 0, 0, 0, 0, 0, 0, 0],
     [0, 0, 0, 0, 0, 0, 0, 0]],
    dtype=int
)

after = np.array(
    [[0, 0, 1, 0, 0, 0, 0, 0],
     [0, 0, 0, 0, 1, 0, 0, 1]],
    dtype=int
)

I want to get a list of indexes of every zero from before table that has been transformed to one in the after table - in this scenario that would be [(0, 2), (1, 4), (1, 7)] .

numpy.ndenumerate is very close to what I'd like to achieve, but it can iterate through only one array at once.

Answer 1

You can pass both conditions to np.logical_and and then use np.argwhere to find indices that meet both conditions:

idx = np.argwhere(np.logical_and(before==0, after==1))

output:

[[0 2]
 [1 4]
 [1 7]]