Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Optimise safety dockerfile #2981

Merged
merged 2 commits into from
May 7, 2023
Merged

Optimise safety dockerfile #2981

merged 2 commits into from
May 7, 2023

Conversation

olliestanley
Copy link
Collaborator

Close #2579. This runs a bit faster on my local machine. Unfortunately I think the main holdup is the PyTorch/CUDA install (also slows down the inference worker image build) which we cannot do much about

@DeanChugall
Copy link
Contributor

Yes, I tried hard to optimize it, but besides Pytorch/Cuda install, I think, also that LAION-AI/blade2blade installation takes the most time.

Aslo copying whole files in command:

COPY --chown="${APP_USER}:${APP_USER}" --from=build /build/lib ${APP_LIBS}.

@olliestanley
Copy link
Collaborator Author

Yes, I tried hard to optimize it, but besides Pytorch/Cuda install, I think, also that LAION-AI/blade2blade installation takes the most time.

Aslo copying whole files in command:

COPY --chown="${APP_USER}:${APP_USER}" --from=build /build/lib ${APP_LIBS}.

I hope blade2blade installation will be faster since yesterday's merge of #2505. Previously pip had to build it from the Git repo but now it is on PyPI so we can pip install directly from there

Copy link
Collaborator

@yk yk left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

thank you

@olliestanley olliestanley merged commit 171690a into main May 7, 2023
@olliestanley olliestanley deleted the optimise-safety-dockerfile branch May 7, 2023 20:06
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Optimize the dockerfile of the inference safety image
3 participants