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

[Question] Is there any straight forward way to retrieve the solution and prediction vector during CV? #1448

Open
chogovadze opened this issue Apr 21, 2022 · 3 comments
Labels
enhancement A new improvement or feature question

Comments

@chogovadze
Copy link

Hello everyone!

I was wondering if there was any way to retrieve the predictions and associated solutions that were used to compute the metrics during training. Specifically, in the case of a 10 fold CV it would correspond to an array containing 10 entries with each entry having in turn the predictions and the labels of that fold.

Thank you for your time in advance!

@eddiebergman
Copy link
Contributor

Hi @chogovadze,

Sorry, there's no straight forward way to do this as far as I'm aware, although @mfeurer might know more. Perhaps in the future but for now a lot of the internals are hidden away and hard to access. We also try to limit memory consumption so keeping the splits in memory is not something we would do (I hope) and I imagine the splits are not written to disk either.

Best,
Eddie

@mfeurer
Copy link
Contributor

mfeurer commented Apr 25, 2022

No, unfortunately, we don't have such information available at the moment. It would be good to have access to this kind of information in the future, though, so I'm marking it as feature.

@mfeurer mfeurer added enhancement A new improvement or feature and removed question labels Apr 25, 2022
@chogovadze
Copy link
Author

@eddiebergman I see. I tried to circumvent this issue using global variables but as expected this does not work. @mfeurer I look forward to it! Thanks again

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement A new improvement or feature question
Projects
None yet
Development

No branches or pull requests

3 participants