-
Notifications
You must be signed in to change notification settings - Fork 2
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
Fix things #14
Fix things #14
Conversation
`--gpu` is not a valid `nvidia-docker` flag
…ablePackage.override` and `ml-ops.overridablePackage.overrideAttrs`
…d in config.pythonPackage when using poetry2nix
}); | ||
}) | ||
ml-ops.common.pythonPackage = { | ||
pipe = [ |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
what change are you making here? Are pipe and override attributes intrinsic to all nix objects? or are they just names you're using
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
nvm. I see that they are just constructs that you've made and are manually handling / passing to pipe etc. etc.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
lib.trivial.pipe
is a nixpkgs function, and ml-ops.common.pythonPackage.pipe
is an option of the arguments passed to lib.trivial.pipe
. The pipe
option introduced by this PR is more general than previous override
and overrideAttrs
options.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This PR fixes a lot of issues preventing us from submitting a job to AKS:
libnvidia-container
compatibility issue by cherry-picking fix: letdevenv
-created OCI images be compatible withlibnvidia-container
cachix/devenv#873pkgs
passed topoetry2nix
, so that dependents of Python also get rebuilt agained the customized python. Without the changes dependent packages of Python will conflicts with the customized python.maxLayers
option to build multi-layer images by cherry-picking feat: addcontainers.<name>.maxLayers
option cachix/devenv#872