Closed theogf closed 2 years ago
@sethaxen Do you have something in mind?
I think it could be as simple as sampling twice with the same rng
(for a very short chain) and check if the samples are exactly the same.
I agree, we should have had that test anyways.
Sorry the test I added was incorrect, it should be fixed now
@sethaxen @xukai92 Is there anything missing from this PR?
It would be nice also to have a test that would have broken before but now passes.