Fixing NOTEs received from the R-hub automatic checks

.Rbuildignore

@@ -1,4 +1,5 @@
 ^.*\.Rproj$
 ^\.Rproj\.user$
 ^doc$
-cran-comments.md
+^cran-comments.md$
+^CRAN-SUBMISSION$

DESCRIPTION

@@ -6,7 +6,7 @@ Version: 0.3.1
 Authors@R: c(person("Agnieszka", "Prochenka-Sołtys", email = "[email protected]", role = c("aut"), comment = "previous maintainer for versions <= 0.2.0"),
    person("Piotr", "Pokarowski", role = c("aut")),
    person("Szymon", "Nowakowski", email = "[email protected]", role = c("aut", "cre"), comment = c(ORCID = "0000-0002-1939-9512")))
-Description: Model selection algorithms for regression and classification, where the predictors can be continuous or categorical and the number of regressors may exceed the number of observations. The selected model consists of a subset of numerical regressors and partitions of levels of factors. Aleksandra Maj-Kańska, Piotr Pokarowski and Agnieszka Prochenka, 2015. Delete or merge regressors for linear model selection. Electronic Journal of Statistics 9(2): 1749-1778. <https://projecteuclid.org/euclid.ejs/1440507392>. Piotr Pokarowski and Jan Mielniczuk, 2015. Combined l1 and greedy l0 penalized least squares for linear model selection. Journal of Machine Learning Research 16(29): 961-992. <http://www.jmlr.org/papers/volume16/pokarowski15a/pokarowski15a.pdf>. Piotr Pokarowski, Wojciech Rejchel, Agnieszka Sołtys, Michał Frej and Jan Mielniczuk, 2022. Improving Lasso for model selection and prediction. Scandinavian Journal of Statistics, 49(2): 831–863. <https://doi.org/10.1111/sjos.12546>.
+Description: Model selection algorithms for regression and classification, where the predictors can be continuous or categorical and the number of regressors may exceed the number of observations. The selected model consists of a subset of numerical regressors and partitions of levels of factors. Aleksandra Maj-Kańska, Piotr Pokarowski and Agnieszka Prochenka, 2015. Delete or merge regressors for linear model selection. Electronic Journal of Statistics 9(2): 1749-1778. <https://projecteuclid.org/euclid.ejs/1440507392>. Piotr Pokarowski and Jan Mielniczuk, 2015. Combined l1 and greedy l0 penalized least squares for linear model selection. Journal of Machine Learning Research 16(29): 961-992. <https://www.jmlr.org/papers/volume16/pokarowski15a/pokarowski15a.pdf>. Piotr Pokarowski, Wojciech Rejchel, Agnieszka Sołtys, Michał Frej and Jan Mielniczuk, 2022. Improving Lasso for model selection and prediction. Scandinavian Journal of Statistics, 49(2): 831–863. <doi:10.1111/sjos.12546>.
 License: GPL-2
 Encoding: UTF-8
 LazyData: true

R/DMR4glm.R

@@ -1,5 +1,5 @@
 DMR4glm <- function(X, y, clust.method, lam){
-    if (class(y) != "factor"){
+    if (!inherits(y, "factor")){
        stop("Error: y should be a factor")
     }
     lev <- levels(factor(y))
@@ -13,7 +13,7 @@ DMR4glm <- function(X, y, clust.method, lam){
               stop("Error: non-conforming data: nrow(X) not equal to length(y)")
     }
     ssd <- apply(X, 2, function(x) length(unique(x)))
-    if (ssd[1] == 1 & (class(X[,1]) == "numeric" | class(X[,1]) == "integer")){
+    if (ssd[1] == 1 & (inherits(X[,1], "numeric") | inherits(X[,1], "integer"))){
        X <- X[,-1, drop = FALSE]
        ssd <- ssd[-1]
     }

R/DMR4lm.R

@@ -6,7 +6,7 @@ DMR4lm <- function(X, y, clust.method, lam){
               stop("Error: non-conforming data: nrow(X) not equal to length(y)")
     }
     ssd <- apply(X, 2, function(x) length(unique(x)))   #number of unique values in each column of X
-    if (ssd[1] == 1 & (class(X[,1]) == "numeric" | class(X[,1]) == "integer")){  # removing the first column in case
+    if (ssd[1] == 1 & (inherits(X[,1], "numeric") | inherits(X[,1], "integer"))){  # removing the first column in case
         # in case is a numeric constant in X
         # i.e. in case it is an Intercept. Other than that, constant columns are NOT allowed
        X <- X[,-1, drop = FALSE]   #drop=FALSE keeps the dimensions of X

R/DMRnet-package.R

@@ -33,8 +33,8 @@
 #'
 #' Aleksandra Maj-Kańska, Piotr Pokarowski and Agnieszka Prochenka, 2015. Delete or merge regressors for linear model selection. Electronic Journal of Statistics 9(2): 1749-1778. \url{https://projecteuclid.org/euclid.ejs/1440507392}
 #'
-#' Piotr Pokarowski and Jan Mielniczuk, 2015. Combined l1 and greedy l0 penalized least squares for linear model selection. Journal of Machine Learning Research 16(29): 961-992. \url{http://www.jmlr.org/papers/volume16/pokarowski15a/pokarowski15a.pdf}
+#' Piotr Pokarowski and Jan Mielniczuk, 2015. Combined l1 and greedy l0 penalized least squares for linear model selection. Journal of Machine Learning Research 16(29): 961-992. \url{https://www.jmlr.org/papers/volume16/pokarowski15a/pokarowski15a.pdf}
 #'
-#' Piotr Pokarowski, Wojciech Rejchel, Agnieszka Sołtys, Michał Frej and Jan Mielniczuk, 2022. Improving Lasso for model selection and prediction. Scandinavian Journal of Statistics, 49(2): 831–863. \url{https://doi.org/10.1111/sjos.12546}
+#' Piotr Pokarowski, Wojciech Rejchel, Agnieszka Sołtys, Michał Frej and Jan Mielniczuk, 2022. Improving Lasso for model selection and prediction. Scandinavian Journal of Statistics, 49(2): 831–863. \doi{10.1111/sjos.12546}
 #'
 NULL

R/DMRnet4glm.R

@@ -1,5 +1,5 @@
 DMRnet4glm <- function(X, y, clust.method, o, nlambda, lam, maxp, lambda) {
-    if (class(y) != "factor"){
+    if (!inherits(y, "factor")){
        stop("Error: y should be a factor")
     }
     lev <- levels(factor(y))
@@ -12,7 +12,7 @@ DMRnet4glm <- function(X, y, clust.method, o, nlambda, lam, maxp, lambda) {
               stop("Error: non-conforming data: nrow(X) not equal to length(y)")
     }
     ssd <- apply(X, 2, function(x) length(unique(x)))
-    if (ssd[1] == 1 & (class(X[,1]) == "numeric" | class(X[,1]) == "integer")){
+    if (ssd[1] == 1 & (inherits(X[,1], "numeric") | inherits(X[,1], "integer"))){
        X <- X[,-1, drop = FALSE]
        ssd <- ssd[-1]
     }

R/DMRnet4lm.R

@@ -4,7 +4,7 @@ DMRnet4lm <- function(X, y, clust.method, o, nlambda, lam, maxp, lambda){
               stop("Error: non-conforming data: nrow(X) not equal to length(y)")
     }
     ssd <- apply(X, 2, function(x) length(unique(x)))
-    if (ssd[1] == 1 & (class(X[,1]) == "numeric" | class(X[,1]) == "integer")){
+    if (ssd[1] == 1 & (inherits(X[,1], "numeric") | inherits(X[,1], "integer"))){
        X <- X[,-1, drop = FALSE]
        ssd <- ssd[-1]
     }

R/SOSnet4glm.R

@@ -1,5 +1,5 @@
 SOSnet4glm <- function(X, y, o, nlambda, lam, interc, maxp, lambda){
-          if (class(y) != "factor"){
+          if (!inherits(y, "factor")){
              stop("Error: y should be a factor")
           }
           lev <- levels(y)

R/coef.DMR.R

@@ -25,7 +25,7 @@ coef.DMR <- function(object, df = NULL, ...){
                          colnames(out) <- paste("df", object$df, sep = "")
                          return(out)
          }
-         if(class(df) != "numeric" & class(df) != "integer"){
+         if(!inherits(df, "numeric") & !inherits(df, "integer")){
                       stop("Error: wrong input type, df should have type numeric")
          }
          out <- object$beta[,ncol(object$beta) - df + 1]

R/cv_GIC_indexed.R

@@ -73,7 +73,7 @@ cv_GIC_indexed <- function(X, y, nfolds, model_function, ...) {
 
         } else{
                 if (family == "binomial"){
-                        if (class(y) != "factor"){
+                        if (!inherits(y, "factor")){
                                 stop("Error: y should be a factor")
                         }
                         lev <- levels(factor(y))

R/cv_MD_indexed.R

@@ -31,7 +31,7 @@ cv_MD_indexed <- function(X, y, nfolds, model_function, ...) {
 
         } else{
                 if (family == "binomial"){
-                        if (class(y) != "factor"){
+                        if (!inherits(y, "factor")){
                                 stop("Error: y should be a factor")
                         }
                         lev <- levels(factor(y))

R/glamer_4glm.R

@@ -1,5 +1,5 @@
 glamer_4glm <- function(X, y, clust.method, nlambda, lam, maxp, lambda){
-  if (class(y) != "factor"){
+  if (!inherits(y, "factor")){
     stop("Error: y should be a factor")
   }
   lev <- levels(factor(y))
@@ -12,7 +12,7 @@ glamer_4glm <- function(X, y, clust.method, nlambda, lam, maxp, lambda){
     stop("Error: non-conforming data: nrow(X) not equal to length(y)")
   }
   ssd <- apply(X, 2, function(x) length(unique(x)))
-  if (ssd[1] == 1 & (class(X[,1]) == "numeric" | class(X[,1]) == "integer")){   #removing the intercept from X, as it will be added later to X.full
+  if (ssd[1] == 1 & (inherits(X[,1], "numeric") | inherits(X[,1], "integer"))){   #removing the intercept from X, as it will be added later to X.full
     X <- X[,-1, drop = FALSE]
     ssd <- ssd[-1]
   }

R/glamer_4lm.R

@@ -5,7 +5,7 @@ glamer_4lm <- function(X, y, clust.method, nlambda, lam, maxp, lambda){
     stop("Error: non-conforming data: nrow(X) not equal to length(y)")
   }
   ssd <- apply(X, 2, function(x) length(unique(x)))
-  if (ssd[1] == 1 & (class(X[,1]) == "numeric" | class(X[,1]) == "integer")){
+  if (ssd[1] == 1 & (inherits(X[,1], "numeric") | inherits(X[,1], "integer"))){
     X <- X[,-1, drop = FALSE]
     ssd <- ssd[-1]
   }

R/predict.DMR.R

@@ -68,7 +68,7 @@ predict.DMR <- function(object, newx, df = NULL, type = "link", unknown.factor.l
          if(ncol(Z) != nrow(object$beta) | is.null(ncol(newx))){
                     stop(paste("Error: non-conforming arrays, newx should be a data frame with ncol equal to", nrow(object$beta)))
          }
-         if(class(df) != "numeric" & class(df) != "integer" & is.null(class(df))){
+         if(!inherits(df, "numeric") & !inherits(df, "integer") & is.null(class(df))){
                       stop("Error: wrong input type, df should have type numeric, integer or NULL")
          }
          if (names(object)[3] == "rss") type = "link"

cran-comments.md

@@ -1,6 +1,9 @@
-## R CMD check results
+## local R CMD check results
 There were no ERRORs, WARNINGs, or NOTEs. 
 
+## R-hub check results
+There was a NOTE related to the maintainer change. As I read over the Internet, the previous maintainer will get an email asking for her consent
+
 ## Downstream dependencies
 There seem to be no downstream dependencies:
 

vignettes/getting-started.Rmd

@@ -189,10 +189,10 @@ Please note that 1 is the value of a target class in the `predict` output.
 
 # References
 
-1. Szymon Nowakowski, Piotr Pokarowski and Wojciech Rejchel. 2021. *Group Lasso Merger for Sparse Prediction with High-Dimensional Categorical Data.* arXiv [stat.ME]. <http://arxiv.org/abs/2112.11114>
+1. Szymon Nowakowski, Piotr Pokarowski and Wojciech Rejchel. 2021. *Group Lasso Merger for Sparse Prediction with High-Dimensional Categorical Data.* arXiv [stat.ME]. <https://arxiv.org/abs/2112.11114>
 1. Aleksandra Maj-Kańska, Piotr Pokarowski and Agnieszka Prochenka, 2015. *Delete or merge regressors for linear model selection.* Electronic Journal of Statistics 9(2): 1749-1778. <https://projecteuclid.org/euclid.ejs/1440507392>
-2. Piotr Pokarowski and Jan Mielniczuk, 2015. *Combined l1 and greedy l0 penalized least squares for linear model selection.* Journal of Machine Learning Research 16(29): 961-992. <http://www.jmlr.org/papers/volume16/pokarowski15a/pokarowski15a.pdf>
-4. Piotr Pokarowski, Wojciech Rejchel, Agnieszka Sołtys, Michał Frej and Jan Mielniczuk, 2022. *Improving Lasso for model selection and prediction.* Scandinavian Journal of Statistics, 49(2): 831–863. <https://doi.org/10.1111/sjos.12546>
+2. Piotr Pokarowski and Jan Mielniczuk, 2015. *Combined l1 and greedy l0 penalized least squares for linear model selection.* Journal of Machine Learning Research 16(29): 961-992. <https://www.jmlr.org/papers/volume16/pokarowski15a/pokarowski15a.pdf>
+4. Piotr Pokarowski, Wojciech Rejchel, Agnieszka Sołtys, Michał Frej and Jan Mielniczuk, 2022. *Improving Lasso for model selection and prediction.* Scandinavian Journal of Statistics, 49(2): 831–863. <doi:10.1111/sjos.12546>
 5. Ludwig Fahrmeir, Rita Künstler, Iris Pigeot, Gerhard Tutz, 2004. Statistik: der Weg zur Datenanalyse. 5. Auflage, Berlin: Springer-Verlag.
 6. Dean P. Foster and Edward I. George, 1994. *The Risk Inflation Criterion for Multiple Regression.* The Annals of Statistics 22 (4): 1947–75.