Repository navigation
feat: LASSO/Elastic Net via coordinate descent - #68
Merged
Merged
Conversation
Native Rust L1/L2-regularized regression along regularization path:
- Coordinate descent with warm starts (fast convergence)
- Standardized features, centered targets
- Log-spaced alpha path from alpha_max to alpha_min
- Converts LASSO coefficients to BTR Individuals (sign → ±1)
- Supports Elastic Net via l1_ratio parameter
Qin2014 benchmark (ternary:prevalence):
LASSO: Test AUC 0.853, k=177, 0.1s (fastest)
SA: Test AUC 0.911, k=48, 0.3s
GA: Test AUC 0.791, k=48, 0.5s
6 tests: soft threshold, coordinate descent (zero/high alpha),
basic run, reproducibility, coefficient conversion
Closes #65
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Native Rust LASSO with warm-start regularization path. 0.1s on Qin2014. 6 tests pass. Closes #65.