I have 2 numpy arrays with the same shape. Now I want to copy all values, except 0, from array 2 to array 1.
array 1:
[1, 1, 1]
[1, 1, 1]
[1, 1, 1]
array 2:
[0, 2, 0]
[4, 0, 0]
[6, 6, 0]
The result should now look like this:
[1, 2, 1]
[4, 1, 1]
[6, 6, 1]
How is this possible in Python?
nonzero will return the indices of an array that is not 0.
idx_nonzero = B.nonzero()
A[idx_nonzero] = B[idx_nonzero]
nonzero is also what numpy.where returns when only a condition is passed in. Thus, equivalently, we can do
idx_nonzero = np.where(B != 0) # (B != 0).nonzero()
A[idx_nonzero] = B[idx_nonzero]
This solution is in-place. If you need to create a new array, see @jp_data_analysis' answer.
np.where supports this. Below solution creates a new array. For an in-place alternative, see @Tai's answer.
A = np.array(
[[1, 1, 1],
[1, 1, 1],
[1, 1, 1]])
B = np.array(
[[0, 2, 0],
[4, 0, 0],
[6, 6, 0]])
C = np.where(B==0, A, B)
# [1, 2, 1]
# [4, 1, 1]
# [6, 6, 1]
