Learn R Programming
⚠️There's a newer version (1.3.8) of this package.Take me there.

sglOptim (version 1.0.122.0)

Sparse group lasso generic optimizer

Description

Fast generic solver for sparse group lasso optimization problems. The loss (objective) function must be defined in a C++ module. This package apply template metaprogramming techniques, therefore -- when compiling the package from source -- a high level of optimization is needed to gain full speed (e.g. for the GCC compiler use -O3). Use of multiple processors for cross validation and subsampling is supported through OpenMP. The Armadillo C++ library is used as the primary linear algebra engine. (The sglOptim package version a.b.c.d is interpreted as follows: a - primary version, b - major updates and fixes, c - source revision as corresponding to R-Forge, d - minor fixes made only to the CRAN branch of the source)

Copy Link

Version

Install

install.packages('sglOptim')

Monthly Downloads

75

Version

1.0.122.0

License

GPL (>= 2)

Maintainer

Martin Vincent

Last Published

March 24th, 2014

Functions in sglOptim (1.0.122.0)

Err

Generic function for computing error rates
compute_error

Helper function for computing error rates
create.sgldata

Create a sgldata object
nmod

Generic function for counting the number of models
features

Generic function for extracting nonzero features (or groups)
parameters.sgl

Extracting nonzero parameters
sgl.standard.config

Standard algorithm configuration
sgl_lambda_sequence

Generic routine for computing a lambda sequence for the regularization path
sgl_predict

Sgl predict
rearrange

Generic rearrange function
sgl_cv

Generic sparse group lasso cross validation using multiple possessors
nmod.sgl

Returns the number of models in a sgl object
coef.sgl

Extracting the nonzero coefficients
rearrange.sgldata

Rearrange sgldata
test.data

Simulated data set
models.sgl

Returns the estimated models (that is the $beta$ matrices)
prepare.args

Generic function for preparing the sgl call arguments
models

Generic function for extracting the fitted models
sgl.algorithm.config

Create a new algorithm configuration
prepare.args.sgldata

Prepare sgl function arguments
sgl_fit

Fit a sparse group lasso regularization path.
features.sgl

Extracting nonzero features
sgl_subsampling

Generic sparse group lasso subsampling procedure
parameters

Generic function for extracting nonzero parameters
sgl_print

Print information about sgl object