Skip to content

Subgroup-specific cross-validation#55

Open
Lauren-EylerDang wants to merge 6 commits into
tlverse:masterfrom
Lauren-EylerDang:master
Open

Subgroup-specific cross-validation#55
Lauren-EylerDang wants to merge 6 commits into
tlverse:masterfrom
Lauren-EylerDang:master

Conversation

@Lauren-EylerDang

Copy link
Copy Markdown

Hello,
I added a function folds_subgroup which allows for subgroup-specific cross-validation. In this case, the validation set is comprised entirely of the subgroup of interest, and the training set consists of the rest of the subgroup of interest and all other subgroups' data. I am using this function for a project where the goal is to train an optimal model for a subgroup while using data from other groups. This is my first pull request, so I apologize if there is anything I did not do correctly! I tried to follow the instructions and am happy to fix anything (eg formatting) as needed. Thank you!
-Lauren

@tdhock

tdhock commented Jun 22, 2026

Copy link
Copy Markdown

Hi Lauren, your code seems related to the SOAK algorithm we recently published, https://onlinelibrary.wiley.com/doi/10.1002/sam.70055
and code is available in mlr3resampling, https://mlr3resampling.netlify.app/doc/soaked

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants