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

[FEATURE] Re-design --no-cache #485

Open
hasan7n opened this issue Aug 29, 2023 · 0 comments
Open

[FEATURE] Re-design --no-cache #485

hasan7n opened this issue Aug 29, 2023 · 0 comments

Comments

@hasan7n
Copy link
Contributor

hasan7n commented Aug 29, 2023

Running medperf benchmark run or medperf run with --no-cache currently doesn't work.

After merging #483 , rerunning an experiment with existing prediction folder will raise an error and ask users to manually delete that folder. --no-cache will simply ignore that there is an existing result but it doesn't ignore that there are existing predictions and hence raises an error.

Before implementing the trivial solution (which is simply overwrite predictions when --no-cache), we should think about:

  • an option to proceed directly to results evaluation when predictions are present?
  • Is it 'safe' to delete predictions with a simple --no-cache flag? Running models may take days. Perhaps it's better to move them to some .cache and provide users a command to clear cache to save storage.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

1 participant