Skip to content

Add ShrunkCovariance Estimator #2306

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

Open
icfaust opened this issue Feb 10, 2025 · 0 comments
Open

Add ShrunkCovariance Estimator #2306

icfaust opened this issue Feb 10, 2025 · 0 comments
Labels
good first issue Good for newcomers hacktoberfest help wanted Open to community contribution

Comments

@icfaust
Copy link
Contributor

icfaust commented Feb 10, 2025

The output of EmpiricalCovariance is regularized by a shrinkage value impacted by the overall mean of the data.
The goal would be to implement this estimator with post-processing changes to the fitted empirical covariance.
This is very similar to the OAS project and would combine into a medium project.
When implemented in python re-using our EmpiricalCovariance estimator, this would be an easy project with a
small time commitment. Implementing the super-computing distributed version using python would only work for
distributed-aware frameworks. Extended goals would make this a hard difficulty, medium commitment project. This
would require implementing the regularization in C++ in oneDAL both for CPU and GPU. Then this must be made
available in Scikit-learn-intelex for making a new estimator. This would hopefully follow the design strategy
used for our Ridge Regression estimator.

https://scikit-learn.org/stable/modules/generated/sklearn.covariance.ShrunkCovariance.html

@icfaust icfaust added help wanted Open to community contribution good first issue Good for newcomers hacktoberfest labels Feb 10, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
good first issue Good for newcomers hacktoberfest help wanted Open to community contribution
Projects
None yet
Development

No branches or pull requests

1 participant