Replace multidimensional array with different array on condition (Python3.x)

Question

I have three multidimensional arrays in Python 3.9 where I would like to replace the values of one array by the values of the second on condition of a third. So:

Array1 = [[[4, 2, 3], [1, 0, 10], [7, 4, 6]],
          [[8, 7, 95], [3, 5, 22], [0, 7, 0]],
          [[5, 0, 60], [5, 230, 70], [6, 76, 30]]]
Array2 = [[[12, 3, 4], [4, 55, 0], [0, 5, 76]],
          [[34, 60, 76], [40, 430, 0], [4, 0, 11]],
          [[2, 34, 0], [1, 0, 0], [76, 4, 77]]]
Mask = [[[0, 1, 0], [1, 0, 0], [1, 0, 0]],
        [[0, 1, 0], [0, 1, 1], [1, 1, 1]],
        [[1, 0, 1], [1, 0, 0], [1, 1, 0]]]

I would like to replace the values in the first array by the values of the second array, if at the same location in the mask array, the value is 1. So the result should be:

Array1 = array([[[  4,   3,   3],
                 [  4,   0,  10],
                 [  0,   4,   6]],

                [[  8,  60,  95],
                 [  3, 430,   0],
                 [  4,   0,  11]],

                [[  2,   0,   0],
                 [  1, 230,  70],
                 [ 76,   4,  30]]])

Is there any way to do is with a quick (numpy) command that doesn't require me to loop over the whole array? In practice the whole array has a size of (1500,2500,4), so that would take quite some processing.

Tnx in advance

Yes they are of exact same length. Marked Answer has solved my problems. Thanks though. — Luca van der knaap
– Luca van der knaap, Commented Oct 25, 2021 at 18:02

Marat · Accepted Answer · 2021-10-25 16:45:53Z

3

np.where does just the thing:

np.where(Mask, Array2, Array1)

Output:

array([[[  4,   3,   3],
        [  4,   0,  10],
        [  0,   4,   6]],

       [[  8,  60,  95],
        [  3, 430,   0],
        [  4,   0,  11]],

       [[  2,   0,   0],
        [  1, 230,  70],
        [ 76,   4,  30]]])

answered Oct 25, 2021 at 16:45

Marat

15.8k3 gold badges44 silver badges53 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

Luca van der knaap Over a year ago

This works great. I was unaware that the numpy.where function could handle a masking array like an array of booleans. Thankyou!

Collectives™ on Stack Overflow

Replace multidimensional array with different array on condition (Python3.x)

1 Answer 1

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related