How to replace values of an array with another array using numpy in Python

Question

I want to place the array B (without loops) on the array A with starting index A[0,0]

A=np.empty((3,3))
A[:] = np.nan
B=np.ones((2,2))

The result should be:

array([[ 1., 1., nan],
 [ 1., 1., nan],
 [ nan, nan, nan]])

I tried numpy.place(arr, mask, vals) and numpy.put(a, ind, v, mode='raise') but I have to find the mask or all indexes.

How to do that?

Answer 1

Assign it where you want using indexing

import numpy as np
A = np.empty((3,3))
a[:] = np.nan
B = np.ones((2,2))
A[:B.shape[0], :B.shape[1]] = B
array([[1.00000000e+000, 1.00000000e+000, nan],
 [1.00000000e+000, 1.00000000e+000, nan],
 [nan, nan, nan]])

• hpaulj: How did you get all those nan in A?
• iacolippo: ops, didn't run it and didn't remember how np.empty works. Fixed it