calebwin/emu

Current Project Status Update?

Opened this issue · 2 comments

Hey there! I was curious what the current project status is, as it hasn't been updated in a few years.

I totally understand if you just ran out of time for the project or something like that, but I was curious what your thoughts on the current code state is. Do you feel like the direction was right, and might just need to be updated to later versions of WGPU or something, or do you feel like there are better options now, or maybe the original plan didn't go as well as immagined?

That kind of stuff. :)

I'm thinking of adding a WGPU-powered backend to burn for machine learning, and Emu seems to have a lot of the right goals, but I'm not sure if it'd be better to use WGPU directly, since there hasn't been a lot of activity here for a while.

Hey @zicklag. Thanks for posting this issue. I'm currently not maintaining/developing Emu further. The current status is that it is broken with some of the updates to Rust and other dependent packages.

I do think the direction is right and this is a good abstraction for WGPU for compute. It does need to be updated which I think can be done without an unreasonable amount of work (since it seems most relevant WGPU changes since last Emu update are API related).

Hope that helps!

Thanks a bunch that's exactly what I was looking for! I don't have any real use for it yet, but I might take another look at Emu one day if I need something like it, even if just for reference.

I'm good closing this issue, or leaving it open if you want to help discoverability in the issue list.