autotune: Faster and more Efficient Lasso (than 'glmnet' and 'scalreg') with Data-Driven Tuning

Fits Lasso paths for high-dimensional regression using coordinate descent with automatic, data-driven tuning of the regularization parameter. The implementation is 10 to 50 times faster than the standard 'glmnet' implementation of Lasso and over 100 times faster than scaled Lasso. It also provides a reliable estimate of the regression noise level. For details of the method, see Sadhukhan, Wilms, Smeekes and Basu (2025) "Autotune: fast, accurate, and automatic tuning parameter selection for Lasso" <doi:10.48550/arXiv.2512.11139>.

Version:	0.1.0
Depends:	R (≥ 2.10)
Imports:	Rcpp (≥ 1.0.13)
LinkingTo:	Rcpp
Suggests:	knitr, rmarkdown, glmnet, AUC, ggplot2, ggExtra, dplyr, tidyr, Matrix
Published:	2026-08-21
DOI:	10.32614/CRAN.package.autotune
Author:	Tathagata Sadhukhan [aut, cre], Ines Wilms [aut], Stephan Smeekes [aut], Sumanta Basu [aut]
Maintainer:	Tathagata Sadhukhan <ts767 at cornell.edu>
License:	GPL-2 | GPL-3 [expanded from: GPL (≥ 2)]
NeedsCompilation:	yes
Materials:	README
CRAN checks:	autotune results

Documentation:

Reference manual:	autotune.html , autotune.pdf
Vignettes:	Autotune Lasso (source, R code)

Downloads:

Package source:	autotune_0.1.0.tar.gz
Windows binaries:	r-devel: not available, r-release: autotune_0.1.0.zip, r-oldrel: autotune_0.1.0.zip
macOS binaries:	r-release (arm64): autotune_0.1.0.tgz, r-oldrel (arm64): autotune_0.1.0.tgz, r-release (x86_64): autotune_0.1.0.tgz, r-oldrel (x86_64): autotune_0.1.0.tgz

Linking:

Please use the canonical form https://CRAN.R-project.org/package=autotune to link to this page.