How do I set random elements in a 2d numpy array without looping?

Question

I have a 2d array of zeros onto which I want to set some random elements to ones. First the rows are selected and then the columns, like this:

>>> A = np.zeros((100, 50))
>>> rows = np.random.choice(100, size = 10, replace = False)
>>> cols = np.random.randint(50, size = 10)
>>> A[rows][cols] = 1
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
IndexError: index 24 is out of bounds for axis 0 with size 10

Of course I can solve the problem with an explicit loop:

>>> for row in rows: A[row][np.random.randint(50)] = 1

But I don't want to. Can what I want to do be accomplished using numpy without explicit looping?

You use the same axis twice. A[rows] returns set of row, so its shape is 10,50. Then you try again to get rows of that slice. — splash58
– splash58, Commented Oct 30, 2019 at 13:48
I don't see the difference to just initiating an array with 0's and 1's. Wouldn't np.random.randint(0, 2, (100, 50)) also solve your problem? — user8408080
– user8408080, Commented Oct 30, 2019 at 13:50
But then randint won't help either, because it could potentially yield the same pair of numbers twice — user8408080
– user8408080, Commented Oct 30, 2019 at 13:56

Fruity Medley · Accepted Answer · 2019-10-30 14:24:26Z

1

Provide all indexes in one set of brackets, like so
A[rows,cols] = 1

answered Oct 30, 2019 at 14:24

Fruity Medley

5643 silver badges17 bronze badges

Sign up to request clarification or add additional context in comments.

Collectives™ on Stack Overflow

How do I set random elements in a 2d numpy array without looping?

1 Answer 1

Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related