How to get maximum elements from a list using numpy.isclose()

Asked Feb 02 '21 at 12:51

Active Feb 02 '21 at 12:51

Viewed 40 times

I simply want to choose the elements from legal_values that have the same index as the maximum value in q_values. So in the example below the max q_value is 2, which has index 1, therefore the corresponding element in legal actions is 1. If there are more than one maximum values I want all of the indices of the maximum values to be returned.

When I try the code below:

legal_actions = [0, 1]
q_values = [1,2]
legal_actions[np.isclose(q_values, np.max(q_values), 1e-8)]

I get the following error:

TypeError: only integer scalar arrays can be converted to a scalar index

asked Feb 02 '21 at 12:51

Bazman

2,058
9
45
65

1

`legal_actions` has to be a numpy array, not a list: `legal_actions = np.array([0, 1])` – Stef Feb 02 '21 at 12:58
1

Why not just `legal_actions[q == q.max()]` or so? You're dealing with integers after all. Also, convert your lists to arrays. – Mad Physicist Feb 02 '21 at 23:00
1

@Stef. Argmax won't help if you want all the indices. – Mad Physicist Feb 02 '21 at 23:01
Google your errors please. – Mad Physicist Feb 02 '21 at 23:03
@MadPhysicist: correct, I didn't carefully read the question. – Stef Feb 03 '21 at 07:38

How to get maximum elements from a list using numpy.isclose()

0 Answers0