R/config.R

In nielsrhansen/lsgl: Multivariate Linear Sparse Group Lasso

#
#     Description of this R script:
#     R routine for creating a lsgl algorithm configuration
#
#     Intended for use with R.
#     Copyright (C) 2014 Martin Vincent
#
#     This program is free software: you can redistribute it and/or modify
#     it under the terms of the GNU General Public License as published by
#     the Free Software Foundation, either version 3 of the License, or
#     (at your option) any later version.
#
#     This program is distributed in the hope that it will be useful,
#     but WITHOUT ANY WARRANTY; without even the implied warranty of
#     MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
#     GNU General Public License for more details.
#
#     You should have received a copy of the GNU General Public License
#     along with this program.  If not, see <http://www.gnu.org/licenses/>
#

#' Create a new algorithm configuration
#'
#' With the exception of \code{verbose} it is not recommended to change any of the default values.
#'
#' @param tolerance_penalized_main_equation_loop tolerance threshold.
#' @param tolerance_penalized_inner_loop_alpha tolerance threshold.
#' @param tolerance_penalized_inner_loop_beta tolerance threshold.
#' @param tolerance_penalized_middel_loop_alpha tolerance threshold.
#' @param tolerance_penalized_outer_loop_alpha tolerance threshold.
#' @param tolerance_penalized_outer_loop_beta tolerance threshold.
#' @param tolerance_penalized_outer_loop_gamma tolerance threshold.
#' @param use_bound_optimization if \code{TRUE} hessian bound check will be used.
#' @param use_stepsize_optimization_in_penalizeed_loop if \code{TRUE} step-size optimization will be used.
#' @param stepsize_opt_penalized_initial_t initial step-size.
#' @param stepsize_opt_penalized_a step-size optimization parameter.
#' @param stepsize_opt_penalized_b step-size optimization parameter.
#' @param max_iterations_outer max iteration of outer loop
#' @param inner_loop_convergence_limit inner loop convergence limit.
#' @param verbose If \code{TRUE} some information, regarding the status of the algorithm, will be printed in the R terminal.
#' @return A configuration.
#' @examples
#'
#' set.seed(100) # This may be removed, it ensures consistency of the daily tests
#'
#' ## Simulate from Y=XB+E, the dimension of Y is N x K, X is N x p, B is p x K
#'
#' N <- 50 #number of samples
#' p <- 50 #number of features
#' K <- 25  #number of groups
#'
#' B<-matrix(sample(c(rep(1,p*K*0.1),rep(0, p*K-as.integer(p*K*0.1)))),nrow=p,ncol=K)
#'
#' X<-matrix(rnorm(N*p,1,1),nrow=N,ncol=p)
#' Y<-X%*%B+matrix(rnorm(N*K,0,1),N,K)
#'
#' # Create configuration
#' config <- lsgl.algorithm.config(verbose = FALSE)
#'
#' lambda<-lsgl::lambda(X,Y, alpha=1, lambda.min=.5, intercept=FALSE, algorithm.config = config)
#'
#' fit <-lsgl::fit(X,Y, alpha=1, lambda = lambda, intercept=FALSE, algorithm.config = config)
#' @author Martin Vincent
#' @export
lsgl.algorithm.config <- function(tolerance_penalized_main_equation_loop = 1e-10,
		tolerance_penalized_inner_loop_alpha = 1e-4,
		tolerance_penalized_inner_loop_beta = 1,
		tolerance_penalized_middel_loop_alpha = 0.01,
		tolerance_penalized_outer_loop_alpha = 0.01,
		tolerance_penalized_outer_loop_beta = 0,
		tolerance_penalized_outer_loop_gamma = 1e-5,
		use_bound_optimization = FALSE,
		use_stepsize_optimization_in_penalizeed_loop = TRUE,
		stepsize_opt_penalized_initial_t = 1,
		stepsize_opt_penalized_a = 0.1,
		stepsize_opt_penalized_b = 0.1,
		max_iterations_outer = 1e4,
		inner_loop_convergence_limit = 1e5,
		verbose = TRUE) {

	config <- list()

	config$tolerance_penalized_main_equation_loop <- tolerance_penalized_main_equation_loop

	config$tolerance_penalized_inner_loop_alpha <- tolerance_penalized_inner_loop_alpha
	config$tolerance_penalized_inner_loop_beta <- tolerance_penalized_inner_loop_beta

	config$tolerance_penalized_middel_loop_alpha <- tolerance_penalized_middel_loop_alpha

	config$tolerance_penalized_outer_loop_alpha <- tolerance_penalized_outer_loop_alpha
	config$tolerance_penalized_outer_loop_beta <- tolerance_penalized_outer_loop_beta
	config$tolerance_penalized_outer_loop_gamma <- tolerance_penalized_outer_loop_gamma

	config$use_bound_optimization <- use_bound_optimization

	config$use_stepsize_optimization_in_penalizeed_loop <- use_stepsize_optimization_in_penalizeed_loop
	config$stepsize_opt_penalized_initial_t <- stepsize_opt_penalized_initial_t
	config$stepsize_opt_penalized_a <- stepsize_opt_penalized_a
	config$stepsize_opt_penalized_b <- stepsize_opt_penalized_b

	config$max_iterations_outer <- as.integer(max_iterations_outer)
	config$inner_loop_convergence_limit <- as.integer(inner_loop_convergence_limit)

	config$verbose <- verbose

	return(config)
}

#' Standard algorithm configuration
#'
#' \code{lsgl.standard.config <- lsgl.algorithm.config()}
#'
#' @author Martin Vicnet
#' @export
lsgl.standard.config <- lsgl.algorithm.config();

#' Featch information about the C side configuration of the package
#' @return list
#'
#' @author Martin Vicnet
#' @useDynLib lsgl, .registration=TRUE
#' @export
lsgl.c.config <- function() {
	.Call("r_pkg_c_config")
}