Open kampelmuehler opened 2 years ago
Hi @kampelmuehler I think it is just changing the types, that's a great idea. You're welcome to make a PR for this. Cheers, Thibault
Though now looking at the date of your issue, you may not be interested in this feature anymore
Thanks, you're right - my interest has dwindled meanwhile
Thanks for the great repo! Can you point me to what needs to be done to enable fp16 compatibility? I suppose it will need some rewriting in the CUDA kernels. Is it just changing the types in the function declarations or more work that needs to go into it?