FastSF: Fast Structural Filtering

An implementation of the fast structural filtering with L0 penalty. It includes an adaptive polynomial estimator by minimizing the least squares error with constraints on the number of breaks in their (k + 1)-st discrete derivative, for a chosen integer k >= 0. It also includes generalized structure sparsity constraint, i.e., graph trend filtering. This package is implemented via the primal dual active set algorithm, which formulates estimates and residuals as primal and dual variables, and utilizes efficient active set selection strategies based on the properties of the primal and dual variables.

Version: 0.1.1
Depends: R (≥ 3.0.0)
Imports: Rcpp (≥ 0.12.10), limSolve
LinkingTo: Rcpp, RcppArmadillo
Published: 2017-07-19
Author: Canhong Wen, Xueqin Wang, Yanhe Shen, Aijun Zhang
Maintainer: Canhong Wen <wencanhong at>
License: GPL-3
NeedsCompilation: yes
CRAN checks: FastSF results


Reference manual: FastSF.pdf
Package source: FastSF_0.1.1.tar.gz
Windows binaries: r-devel:, r-release:, r-oldrel:
OS X El Capitan binaries: r-release: FastSF_0.1.1.tgz
OS X Mavericks binaries: r-oldrel: FastSF_0.1.1.tgz
Old sources: FastSF archive


Please use the canonical form to link to this page.