How to delete an element in an array depending on another array modification in Python

This may be what you are looking for:

x_0 = np.where(x==0)[0]

rand_ind = np.random.randint(1,len(x_0))

x[x_0[rand_ind]] = 1;

y = np.delete(y, rand_ind)

x_0 is an array of elements in x initially equal to 0. rand_ind is a random index chosen from x_0 length range. To change 0 to 1 the program uses x_0[rand_ind] while to delete the corresponding element in y it simply uses rand_ind.

This is a modified solution for the updated question,

import numpy as np
import random

x_0 = np.where(x==0)[0]

rand_ind = np.array(random.sample(list(x_0), 3))

x[rand_ind] = 1

y = np.delete(y, [np.where(x_0==i)[0] for i in rand_ind])

In this solution rand_ind is a direct choice of indices of the x array and not the trimmed x_0 array. It chooses the indices stored in x_0 to avoid choosing indexes where the value in x is 1. The indexes i.e. elements of x_0 are the indexes of x while the indexes of x_0 correspond to indexes/positions in y.

Program uses random.sample as a straightforward way to pick N elements at random without repetition (here 3). It then turns the elements at rand_ind in x to 1 and deletes the corresponding elements in y. The last step is done by using a generator and finding the indices of x_0 elements with each rand_ind value.