picasso: Sparse Learning with Convex and Concave Penalties

Fast regularization paths for sparse Gaussian, binomial, Poisson, square-root-lasso, and multinomial models with lasso, smoothly clipped absolute deviation, or minimax concave penalties. Computation uses pathwise coordinate optimization, active-set updates, warm starts, screening rules, Proximal Newton iterations, quadratic majorization, and adaptive local linear approximation where appropriate. Core solvers are implemented in C++, and coefficient paths are returned as Matrix-compatible objects.

Version: 2.0.1
Depends: R (≥ 3.1.0), Matrix
Imports: methods
Suggests: testthat (≥ 3.0.0)
Published: 2026-07-25
DOI: 10.32614/CRAN.package.picasso
Author: Jason Ge [aut], Xingguo Li [aut], Haoming Jiang [aut], Mengdi Wang [aut], Tong Zhang [aut], Han Liu [aut], Tuo Zhao [aut, cre], Gael Guennebaud [ctb] (Contributor to bundled Eigen headers), Benoit Jacob [ctb] (Contributor to bundled Eigen headers), Eigen Library Authors [cph] (Copyright holders of bundled Eigen headers in src/include/eigen3)
Maintainer: Tuo Zhao <tourzhao at gatech.edu>
License: GPL-3
Copyright: See inst/COPYRIGHTS for bundled third-party copyright and license notices.
picasso copyright details
NeedsCompilation: yes
Materials: NEWS
CRAN checks: picasso results

Documentation:

Reference manual: picasso.html , picasso.pdf
Vignettes: PICASSO 2.0.0 user guide (source)

Downloads:

Package source: picasso_2.0.1.tar.gz
Windows binaries: r-devel: picasso_1.5.zip, r-release: picasso_1.5.zip, r-oldrel: picasso_1.5.zip
macOS binaries: r-release (arm64): picasso_2.0.1.tgz, r-oldrel (arm64): picasso_2.0.1.tgz, r-release (x86_64): picasso_2.0.1.tgz, r-oldrel (x86_64): picasso_2.0.1.tgz
Old sources: picasso archive

Linking:

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