Filling in a numpy ndarray with a 1d array and a mask/index of true/false values

Let’s say I have the following:

import numpy as np
X = np.array([[1, 2, 3], [4, 5, 6]])
mask = np.array([[False, False, True], [False, True, False]])
values = np.array([20, 30])

I want to fill in the values of 20 and 30 in where the mask is true row wise so the result is:
[[1, 2, 20], [4, 30, 6]].

Also, how would I fill in the vector column wise so the result would be:
[[1, 2, 30], [4, 20, 6]].

How would I do this with just numpy?

>Solution :

X[mask] = values. This works when X.shape == mask.shape and len(values) == mask.sum().

Output:

array([[ 1,  2, 20],
       [ 4, 30,  6]])

For the second part of your question, you can transpose X and mask for the computation:

X.T[mask.T] = values

Output:

array([[ 1,  2, 30],
       [ 4, 20,  6]])

See the numpy documentation