After much messing around with the Windows virtual environment, I’ve gotten it to the point where torch.cuda.is_available() returns True.
But now I have a totally different problem: calling ray.init() (and my original script) outputs this line:
2021-07-02 12:41:47,233 INFO services.py:1274 -- View the Ray dashboard at http://127.0.0.1:8265
then hangs.
I’m using ray 1.4.1
Created a new discussion: Ray.init() hangs