tristandeleu/pytorch-maml-rl

Cuda Support Issue

Opened this issue · 2 comments

imhgchoi commented

Hi,
Thank you so much for the codes.

I see that cuda may not be supported in many cases.
I'm guessing this is due to multiprocessing, which I am not so sure about.
So can you (or whoever can) briefly explain what's causing it, and whether it can be fixed?

Thanks in advance :)

tristandeleu commented

Multiprocessing is indeed an issue, and that probably requires changing the context to spawn. However there are some parts of the code which might not work right away with spawn context (see #51).

I haven't used this code on GPU so I don't know what would be required to fix it exactly, but maybe @lucifer2859 has some input on that (if you were actually trying to run it on GPU?)?

imhgchoi commented

I see. Well, actually I'm running the code from mp-spawn branch.
Unfortunately however, cuda doesn't work with it as well.

Anyways, I really appreciate your immediate response! Thank you :)