They are busy doing ML research and waiting to be given a recipe to follow. That, and ML researchers can't also be expected to be good at everything. People also tend to forget that these ML packages are ridiculously complicated, and have a lot of dependencies not just on other libraries but on particularities of your system. Exciting!Īs if compiling any given C program wasn't also a crap shoot. I've reached the limits of my knowledge on this, but will following closely as new PRs are merged in over the coming days. This is apparently a known issue and a fix is being tested. Sometimes it rendered images just as a black canvas, other times it worked. It kept throwing the "leaked semaphor objects" error someone else reported (even when rendering at 64圆4). Not sure if that's a coincidence, or if they've included extra optimisations. Running txt2img.py from lstein's repo seems to run about 30% faster than OP's. It offers better UX for interacting with Stable Diffusion, and seems to be a promising project. ĮDIT 2: After playing around with this repo, I've found: Follow the main installation instructions here. Install the other build requirements referenced in OP's setup:ģ. Install `conda`, if you don't already have it:Ģ. `rm` the existing `stable-diffusion` repo (assuming you followed OP's original setup)ġ. Giving it a try now.ĮDIT: Got it working, with a couple of pre-requisite steps:Ġ. Brilliant, thank you! I just got OP's setup working, but this seems much more user-friendly.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |