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

Do not instantiate parts of the system that are unnecessary when performing a stage #3

Open
ibro45 opened this issue Apr 14, 2022 · 3 comments
Assignees
Labels
enhancement New feature or request

Comments

@ibro45
Copy link
Collaborator

ibro45 commented Apr 14, 2022

When in val, test, or predict mode, optimizers and schedulers aren't necessary.

When running training, no need for test or predict parts. And so on.

@surajpaib
Copy link
Collaborator

Revisiting this, I think we can easily do it with the _disabled_ flag now in the config

@ibro45
Copy link
Collaborator Author

ibro45 commented Apr 22, 2023

I had tried to do it a month ago or so, but was running into the problem of referencing/interpolating config values from the disabled parts if I remember correctly.

Also, with the latest monai version, % referencing seems to work even less than it did. I should really open an issue about why % (referencing a config item as-is) is not as usable as @ (referencing a config item that is resolved, e.g., instantiated or code executed).

Mentioning this because you should be able to %-reference a disabled config item. Obviously, @-referencing should not work for the disabled config item because it's not being resolved.

@ibro45 ibro45 added the enhancement New feature or request label Oct 10, 2023
@ibro45
Copy link
Collaborator Author

ibro45 commented Oct 10, 2023

See if this can be handled nicely when we'll be switching to the BundleWorkflows

@ibro45 ibro45 changed the title Do not instantiate parts of the system that are unnecessary when performing val/test Do not instantiate parts of the system that are unnecessary when performing a stage Mar 28, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement New feature or request
Projects
None yet
Development

No branches or pull requests

2 participants