AbdBarho

Meeting notes Inference System: * Separate the inference from the website * uses its own auth system * Email may not be feasible. Preprompt: * Language selection should be available,...

this is mainly because storybook does not support react 18, and had to be installed with legacy peer deps

@ParisNeo Has this solution been discussed in an issue with the team members before? Our plan was to collect manually created datasets to fine tune a model, I am not...

@davedavis pretty much what you said I did some tweaks to the xformers build so that it targets all supported cuda architectures to accumudate the hardware variation for users of...

@bitshifter52 what are your versions of nvcc, torch, and python?

@webdevbyjoss maybe it is a typo but: ``` docker-compose ``` and ``` docker compose ``` are two different things also, `--profile [x] up`, not `up --profile [x]`

Overall I think its a good plan, some comments: > (2) pip install xformers downloads a source distribution. At setup, we retrieve the pytorch/cuda/xformers/python/os version, and download the corresponding binary...

@danthe3rd now that we have pypi storage, what is the next step? is there something I can help with? do we still need to worry about the pip index? it...

@danthe3rd happy thanksgiving! ping me when you have any updates.

for 1) we are already building on manylinux https://github.com/facebookresearch/xformers/blob/1515f775739d89628f0293aadcdc3a820d43f287/.github/workflows/wheels.yml#L65 but the generated wheel has the suffix `linux_x64_86`, maybe we can cheese it and `sed` to `manylinux`, I used the 2014...