Improve R package compliance with R CMD check and enhance graph rendering compatibility

DESCRIPTION

@@ -7,14 +7,14 @@ Authors@R:
            role = c("aut", "cre"),
            email = "genta.kik@gmail.com")
 Description: R implementation of LiNGAM algorithms based on Python implementation (The LiNGAM Project: https://sites.google.com/site/sshimizu06/lingam).
-License: GPL-3
+License: MIT + file LICENSE
 Encoding: UTF-8
-LazyData: true
 Roxygen: list(markdown = TRUE)
-RoxygenNote: 7.1.2
+RoxygenNote: 7.3.2
 Imports: 
     lars,
     DiagrammeR,
     glmnet,
     clue,
-    fastICA
+    fastICA,
+    stats

LICENSE

@@ -1,2 +1,2 @@
-YEAR: 2020
+YEAR: 2025
 COPYRIGHT HOLDER: G. Kikuchi

NAMESPACE

@@ -8,4 +8,5 @@ export(plot_adjacency_mat)
 import(DiagrammeR)
 importFrom(clue,solve_LSAP)
 importFrom(fastICA,fastICA)
+importFrom(glmnet,cv.glmnet)
 importFrom(lars,lars)

R/BaseLiNGAM.R

@@ -6,6 +6,7 @@
 #' This code is based on Python implementation:
 #'
 #' (The LiNGAM Project: https://sites.google.com/site/sshimizu06/lingam)
+#' @importFrom glmnet cv.glmnet
 #' @importFrom lars lars
 #' @export
 BaseLiNGAM <- R6::R6Class("LiNGAM",

R/ICALiNGAM.R

@@ -31,17 +31,17 @@ ICALiNGAM <- R6::R6Class("ICALiNGAM", inherit = BaseLiNGAM,
         fit = function(X) {
 
             # obtain unmixing matrix W_ica
-            ica <- fastICA(X,
-                           n.comp  = ncol(X),
-                           alg.typ = "parallel",
-                           maxit   = self$max_iter,
-                           fun     = "logcosh"
+            ica <- fastICA::fastICA(X,
+                                    n.comp  = ncol(X),
+                                    alg.typ = "parallel",
+                                    maxit   = self$max_iter,
+                                    fun     = "logcosh"
             )
             W_ica <- (ica$K %*% ica$W)
 
             # obtain permuted W_ica
             # note that W_ica is t(W_ica) of scipy's fastICA
-            linassign <- solve_LSAP(1 / abs(W_ica), maximum = FALSE)
+            linassign <- clue::solve_LSAP(1 / abs(W_ica), maximum = FALSE)
             col_index <- unname(linassign)
             PW_ica    <- t(W_ica[, col_index])
 

R/gen_dummy_data.R

@@ -6,12 +6,12 @@
 #' @export
 gen_dummy_data <- function(num_samples = 10000, random_state = 1) {
     set.seed(random_state)
-    x3 <- runif(n = num_samples)
-    x0 <- runif(n = num_samples) + 3 * x3
-    x2 <- runif(n = num_samples) + 6 * x3
-    x1 <- runif(n = num_samples) + 3 * x0 + 2 * x2
-    x5 <- runif(n = num_samples) + 4 * x0
-    x4 <- runif(n = num_samples) + 8 * x0 - 1 * x2
+    x3 <- stats::runif(n = num_samples)
+    x0 <- stats::runif(n = num_samples) + 3 * x3
+    x2 <- stats::runif(n = num_samples) + 6 * x3
+    x1 <- stats::runif(n = num_samples) + 3 * x0 + 2 * x2
+    x5 <- stats::runif(n = num_samples) + 4 * x0
+    x4 <- stats::runif(n = num_samples) + 8 * x0 - 1 * x2
     X  <- data.frame(x0, x1, x2, x3, x4, x5)
     return(X)
 }

R/plot_adjacency_mat.R

@@ -27,7 +27,6 @@ plot_adjacency_mat <- function(adj_mat,
 
         }
     }
-    dot <- DiagrammeR::render_graph(graph, layout = "dot", title = title)
-    dot$x$diagram <- gsub("neato", "dot", dot$x$diagram)
-    print(DiagrammeR::grViz(dot$x$diagram))
+
+    DiagrammeR::render_graph(graph, layout = "neato", title = title)
 }