-
Notifications
You must be signed in to change notification settings - Fork 99
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
Use multi stage docker build #15
Labels
enhancement
New feature or request
Comments
How about you open up a pull request with your current changes and push an image to docker hub. I would try to find some people who use CPUs that don't support AVX, to install and test it. |
Created PR #17 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Suggestion: Use multi stage docker build to only copy data model files instead of also including the whole tar file.
Multi stage build will also mean that only the dlib binaries are packaged; the dlib source and dlib object files won't be part of the build. Preliminary testing on my system shows an image reduction about ~1GB on the backend device. If any other deadweight from the build process can also be shed, it can make the image much leaner and easier to run on lower end devices.
I would submit a pull request but I'm not certain how the dlib binaries are installed/referred, so just a direct change to the dockerfile may cause regressions.
The text was updated successfully, but these errors were encountered: