Merge branch 'master' of https://github.com/KrystynaGrzesiak/graphslope

gslope/NAMESPACE

@@ -1,6 +1,8 @@
 # Generated by roxygen2: do not edit by hand
 
 S3method(coef,gslope)
+S3method(graph_plot,gslope)
+S3method(plot,gslope)
 S3method(summary,gslope)
 export(ADMM_algorithm)
 export(coef)

gslope/R/admm_algorithm.R

@@ -16,14 +16,16 @@ prox_matrix = function(matrix_in, lambda) {
 #' @description  Executes ADMM algorithm to find precision matrix.
 #' @param sample_cov variance-covariance matrix.
 #' @param lambda vector of SLOPE regularizers.
-#' @param mu correction for scaling lambda.
+#' @param mu correction for lambda scaling.
 #' @param max_iter maximum number of iterations of ADMM algorithm. Default 10 000.
 #' @param epsilon a value determining stop condition.
-#' @keywords TODO: keywords
-#' @return \code{ADMM_algorithm} A list with components:
-#' \item{precision_matrix}{precision matrix revealing graph structure for the data.}
-#' \item{iterations}{number of iterations before stop condition or \code{max_iter}.}
-#' @details #TODO: details
+#' @keywords admm
+#' @return \code{ADMM_algorithm} returns a list with components:
+#' \itemize{
+#' \item \code{precision_matrix} precision matrix revealing graph structure for the data.
+#' \item \code{iterations} number of iterations before stop condition or \code{max_iter}.
+#' }
+#' @details ADMM algorithm is used to solve convex optimization problems by dividing them into smaller, easier to solve problems.
 #' @examples
 #' sample_cov = cov(scale(mtcars))
 #' lambda = gslope::create_lambda(sample_cov, nrow(mtcars))

gslope/R/create_lambda.R

@@ -6,7 +6,7 @@
 #' @keywords lambda, graphical slope, penalty
 #' @return \code{create_lambda()} returns a vector of the length \eqn{p * (p-1)/2}, which is equal
 #' to number of elements in lower (upper) triangle of a variance-covariance matrix excluding diagonal.
-#' @details The Lambda series are computed based on Benjamini-Hochberg's correction.
+#' @details The Lambda series are computed based on Benjamini-Hochberg's procedure.
 #' @examples
 #' sample_cov = cov(scale(mtcars))
 #' create_lambda(sample_cov, nrow(mtcars))

gslope/R/summary.R

@@ -3,12 +3,22 @@
 #'
 #' @param x an object of class `'gslope'`
 #'
-#' @return Prints output on the screen
+#' @return Prints following output on the screen:
+#' \itemize{
 #'
+#' \item \code{call} formula used in function call.
+#' \item \code{precision matrix} a sparse precision matrix
+#' \item \code{covariance matrix} covariance matrix for the data
+#' \item \code{scaled precision matrix} see: ?gslope
+#' \item \code{lambda} lambda regularizers used
+#' \item \code{iterations} number of iterations
+#' \item \code{graph} a graph object from igraph package. For more details see ?igraph.
+#' \item \code{clusters} clusters revealing sub-graph structure in the model For more details see ?gslope.
+#' }
 #' @examples
 #' d = gslope(as.matrix(mtcars))
 #' summary(d)
-#' 
+#'
 #' @export
 summary <- function(x){
   UseMethod("summary")