Package: fuser 1.0.1

Frank Dondelinger

fuser: Fused Lasso for High-Dimensional Regression over Groups

Enables high-dimensional penalized regression across heterogeneous subgroups. Fusion penalties are used to share information about the linear parameters across subgroups. The underlying model is described in detail in Dondelinger and Mukherjee (2017) <arxiv:1611.00953>.

Authors:Frank Dondelinger [aut, cre], Olivier Wilkinson [aut]

fuser_1.0.1.tar.gz
fuser_1.0.1.tar.gz(r-4.5-noble)fuser_1.0.1.tar.gz(r-4.4-noble)
fuser_1.0.1.tgz(r-4.4-emscripten)fuser_1.0.1.tgz(r-4.3-emscripten)
fuser.pdf |fuser.html✨
fuser/json (API)
NEWS

# Install 'fuser' in R:
install.packages('fuser', repos = 'https://cloud.r-project.org')
Uses libs:
  • c++– GNU Standard C++ Library v3

On CRAN:

Conda:

This package does not link to any Github/Gitlab/R-forge repository. No issue tracker or development information is available.

cpp

2.70 score 181 downloads 5 exports 12 dependencies

Last updated 7 years agofrom:6cd4856235. Checks:3 OK. Indexed: yes.

TargetResultLatest binary
Doc / VignettesOKMar 11 2025
R-4.5-linux-x86_64OKMar 11 2025
R-4.4-linux-x86_64OKMar 11 2025

Exports:bigeigenfusedL2DescentGLMNetfusedLassoProximalfusedLassoProximalIterationsTakengenerateBlockDiagonalMatrices

Dependencies:codetoolsforeachglmnetirlbaiteratorslatticeMatrixRcppRcppEigenRSpectrashapesurvival

Using the fuser package for prediction over subgroups

Frank Dondelinger

Rendered fromsubgroup_fusion.Rmdusingknitr::rmarkdownon Mar 11 2025.

Last update: 2017-08-17
Started: 2017-08-17

Citation

To cite package ‘fuser’ in publications use:

Dondelinger F, Wilkinson O (2018). fuser: Fused Lasso for High-Dimensional Regression over Groups. R package version 1.0.1, https://CRAN.R-project.org/package=fuser.

Corresponding BibTeX entry: 

  @Manual{,
    title = {fuser: Fused Lasso for High-Dimensional Regression over
      Groups},
    author = {Frank Dondelinger and Olivier Wilkinson},
    year = {2018},
    note = {R package version 1.0.1},
    url = {https://CRAN.R-project.org/package=fuser},
  }

Readme and manuals

fuser

Fused lasso for high-dimensional regression over groups. This package implements the model described in Dondelinger et al. (2016).

Installation

library('devtools')
install_github('FrankD/fuser')

Example

See also the included vignette.

library(fuser)
set.seed(123)

# Generate simple heterogeneous dataset
k = 4 # number of groups
p = 100 # number of covariates
n.group = 15 # number of samples per group
sigma = 0.05 # observation noise sd
groups = rep(1:k, each=n.group) # group indicators

# sparse linear coefficients
beta = matrix(0, p, k)
nonzero.ind = rbinom(p*k, 1, 0.025/k) # Independent coefficients
nonzero.shared = rbinom(p, 1, 0.025) # shared coefficients
beta[which(nonzero.ind==1)] = rnorm(sum(nonzero.ind), 1, 0.25) 
beta[which(nonzero.shared==1),] = rnorm(sum(nonzero.shared), -1, 0.25)

X = lapply(1:k, function(k.i) matrix(rnorm(n.group*p),n.group, p)) # covariates 
y = sapply(1:k, function(k.i) X[[k.i]] %*% beta[,k.i] + rnorm(n.group, 0, sigma)) # response
X = do.call('rbind', X)

# Pairwise Fusion strength hyperparameters (tau(k,k'))
# Same for all pairs in this example
G = matrix(1, k, k) 

# Use L1 fusion to estimate betas (with near-optimal sparsity and 
# information sharing among groups)
beta.estimate = fusedLassoProximal(X, y, groups, lambda=0.001, tol=9e-5, 
                                   gamma=0.001, G, intercept=FALSE,
                                   num.it=2000) 

# Generate block diagonal matrices for L2 fusion approach
transformed.data = generateBlockDiagonalMatrices(X, y, groups, G)

# Use L2 fusion to estimate betas (with near-optimal information sharing among groups)
beta.estimate = fusedL2DescentGLMNet(transformed.data$X, transformed.data$X.fused, 
                                     transformed.data$Y, groups, lambda=c(0,0.001,0.1,1),
                                     gamma=0.001)