Replace some elements of numpy.ndarray with zero given another numpy array

Question

I'd need to know the most efficient way for the following case. There is a numpy.ndarray of shape 11k*11k for which I need to force all elements of some rows to be zero given a binary numpy array of shape 11k. A toy example could be described as follows:

Inputs:

x = np.array([[2, 1, 1, 2],
              [0, 2, 1, 0],
              [1, 0, 1, 1],
              [2, 2, 1, 0]])

ref = np.array([0, 1, 1, 0])

Output:

y = ([[0, 0, 0, 0],
       [0, 2, 1, 0],
       [1, 0, 1, 1],
       [0, 0, 0, 0]])

Extend `ref` to 2D and multiply - `x*ref[:,None]`. – Divakar Jun 17 '19 at 16:57 — Divakar, Jun 17 '19 at 16:57

score 2 · Accepted Answer · answered Jun 17 '19 at 17:02

2

Use this -

y = np.multiply(x.T,ref).T

array([[0, 0, 0, 0],
       [0, 2, 1, 0],
       [1, 0, 1, 1],
       [0, 0, 0, 0]])

answered Jun 17 '19 at 17:02

Akshay Sehgal

18,741
3
21
51

Replace some elements of numpy.ndarray with zero given another numpy array

1 Answers1