How to set a probability of a value becoming a zero for an np.array?

Question

I've got an np.array 219 by 219 with mostly 0s and 2% of nonzeros and I know want to create new arrays where each of the nonzero values has 90% of chance of becoming a zero.

I now know how to change the n-th non zero value to 0 but how to work with probabilities?

Probably this can be modified:

index=0
for x in range(0, 219):
    for y in range(0, 219):
        if (index+1) % 10 == 0:
            B[x][y] = 0
        index+=1
print(B)

Answer 1

You could use np.random.random to create an array of random numbers to compare with 0.9, and then use np.where to select either the original value or 0. Since each draw is independent, it doesn't matter if we replace a 0 with a 0, so we don't need to treat zero and nonzero values differently. For example:

In [184]: A = np.random.randint(0, 2, (8,8))

In [185]: A
Out[185]: 
array([[1, 1, 1, 0, 0, 0, 0, 1],
       [1, 1, 1, 0, 0, 0, 0, 0],
       [1, 1, 1, 1, 1, 0, 0, 0],
       [0, 1, 0, 1, 0, 0, 0, 1],
       [0, 1, 0, 1, 1, 1, 1, 0],
       [1, 1, 0, 1, 1, 0, 0, 0],
       [1, 0, 0, 1, 0, 0, 1, 0],
       [1, 1, 0, 0, 0, 1, 0, 1]])

In [186]: np.where(np.random.random(A.shape) < 0.9, 0, A)
Out[186]: 
array([[0, 0, 0, 0, 0, 0, 0, 0],
       [0, 0, 0, 0, 0, 0, 0, 0],
       [0, 0, 0, 0, 0, 0, 0, 0],
       [0, 0, 0, 0, 0, 0, 0, 1],
       [0, 0, 0, 0, 0, 0, 0, 0],
       [0, 0, 0, 0, 0, 0, 0, 0],
       [0, 0, 0, 0, 0, 0, 0, 0],
       [0, 1, 0, 0, 0, 0, 0, 0]])