Fixing roxygen issues in RMD Check

pkg/caret/NAMESPACE

@@ -65,6 +65,7 @@ S3method(oob_pred,sbf)
 S3method(oob_pred,train)
 S3method(parallelplot,resamples)
 S3method(pcaNNet,formula)
+S3method(plot,prcomp.resamples)
 S3method(plsda,default)
 S3method(posPredValue,default)
 S3method(posPredValue,matrix)
@@ -109,7 +110,6 @@ S3method(print,confusionMatrix.rfe)
 S3method(print,confusionMatrix.sbf)
 S3method(print,confusionMatrix.train)
 S3method(print,diff.resamples)
-S3method(print,dummyVars)
 S3method(print,expoTrans)
 S3method(print,gafs)
 S3method(print,icr)
@@ -153,9 +153,12 @@ S3method(splsda,default)
 S3method(stripplot,rfe)
 S3method(stripplot,train)
 S3method(summary,bag)
+S3method(summary,diff.resamples)
+S3method(summary,resamples)
 S3method(train,default)
 S3method(train,formula)
 S3method(trim,train)
+S3method(update,rfe)
 S3method(varImp,C5.0)
 S3method(varImp,JRip)
 S3method(varImp,PART)
@@ -195,8 +198,6 @@ export(F_meas)
 export(LPH07_1)
 export(LPH07_2)
 export(MeanSD)
-export(R2)
-export(RMSE)
 export(SLC14_1)
 export(SLC14_2)
 export(anovaScores)
@@ -305,7 +306,6 @@ export(pickVars)
 export(plot.calibration)
 export(plot.gafs)
 export(plot.lift)
-export(plot.prcomp.resamples)
 export(plot.rfe)
 export(plot.safs)
 export(plot.train)
@@ -358,8 +358,6 @@ export(splsda)
 export(sumDiss)
 export(summary.bagEarth)
 export(summary.bagFDA)
-export(summary.diff.resamples)
-export(summary.resamples)
 export(summary.train)
 export(svmBag)
 export(tolerance)
@@ -373,7 +371,6 @@ export(twoClassSim)
 export(twoClassSummary)
 export(upSample)
 export(update.gafs)
-export(update.rfe)
 export(varImp)
 export(var_seq)
 export(well_numbered)

pkg/caret/R/aaa.R

@@ -221,15 +221,14 @@ best <- function(x, metric, maximize)
   bestIter
 }
 
-
 #' @export
 defaultSummary <- function(data, lev = NULL, model = NULL)
 {
   if(is.character(data$obs)) data$obs <- factor(data$obs, levels = lev)
   postResample(data[,"pred"], data[,"obs"])
 }
 
-
+#' @rdname postResample
 #' @importFrom ModelMetrics auc
 #' @export
 twoClassSummary <- function (data, lev = NULL, model = NULL)
@@ -250,6 +249,7 @@ twoClassSummary <- function (data, lev = NULL, model = NULL)
   out
 }
 
+#' @rdname postResample
 #' @importFrom stats complete.cases
 #' @export
 mnLogLoss <- function(data, lev = NULL, model = NULL){
@@ -266,6 +266,7 @@ mnLogLoss <- function(data, lev = NULL, model = NULL){
   ModelMetrics::mlogLoss(dataComplete$obs, probs)
 }
 
+#' @rdname postResample
 #' @export
 multiClassSummary <- function (data, lev = NULL, model = NULL){
   #Check data

pkg/caret/R/bagEarth.R

@@ -298,6 +298,7 @@ print.bagEarth <- function (x, ...)
 #' summary(fit)
 #' }
 #'
+#' @method summary bagEarth
 #' @export summary.bagEarth
 "summary.bagEarth" <-
   function(object, ...)

pkg/caret/R/calibration.R

@@ -17,7 +17,7 @@
 #'
 #' @param lattice.options A list that could be supplied to \code{\link[lattice:lattice.options]{lattice#' .options}}
 #'
-#' @param bwidth, dwidth a numeric value for the confidence interval bar width and dodge width, respective#' ly. In the latter case, a dodge is only used when multiple models are specified in the formula.
+#' @param bwidth,dwidth a numeric value for the confidence interval bar width and dodge width, respective#' ly. In the latter case, a dodge is only used when multiple models are specified in the formula.
 #' @param \dots options to pass through to \code{\link[lattice:xyplot]{xyplot}} or the panel function (not #' used in \code{calibration.formula}).
 #'
 #' @details

pkg/caret/R/dummyVar.R

@@ -1,42 +1,3 @@
-#' @export
-contr.ltfr <- function (n, contrasts = TRUE, sparse = FALSE)
-{
-  if (is.numeric(n) && length(n) == 1L) {
-    if (n > 1L)
-      levels <- as.character(seq_len(n))
-    else stop("not enough degrees of freedom to define contrasts")
-  }
-  else {
-    levels <- as.character(n)
-    n <- length(n)
-  }
-  contr <- .RDiag(levels, sparse = sparse)
-  if (contrasts) {
-    if (n < 2L) stop(gettextf("contrasts not defined for %d degrees of freedom", n - 1L), domain = NA)
-  }
-  contr
-}
-
-#' @export
-contr.dummy <- function(n, ...)
-{
-  if (is.numeric(n) && length(n) == 1L) {
-    if (n > 1L)
-      levels <- as.character(seq_len(n))
-    else stop("not enough degrees of freedom to define contrasts")
-  }
-  else {
-    levels <- as.character(n)
-    n <- length(n)
-  }
-  out <- diag(n)
-  rownames(out) <- levels
-  colnames(out) <- levels
-  out
-}
-
-
-
 #' Create A Full Set of Dummy Variables
 #'
 #' \code{dummyVars} creates a full set of dummy variables (i.e. less than full
@@ -88,9 +49,6 @@ contr.dummy <- function(n, ...)
 #' @param contrasts A logical indicating whether contrasts should be computed.
 #' @param sparse A logical indicating if the result should be sparse.
 #' @param x A factor vector.
-#' @param drop2nd A logical: when the factor \code{x} has two levels, should
-#' both dummy variables be returned (\code{drop2nd = FALSE} or only the dummy
-#' variable for the first level \code{drop2nd = TRUE}.
 #' @param ... additional arguments to be passed to other methods
 #' @return The output of \code{dummyVars} is a list of class 'dummyVars' with
 #' elements \item{call }{the function call} \item{form }{the model formula}
@@ -205,6 +163,8 @@ dummyVars.default <- function (formula, data, sep = ".", levelsOnly = FALSE, ful
 
 }
 
+#' @rdname dummyVars
+#' @method predict dummyVars
 #' @export
 print.dummyVars <- function(x, ...)
 {
@@ -269,4 +229,42 @@ predict.dummyVars <- function(object, newdata, na.action = na.pass, ...)
   x[, colnames(x) != "(Intercept)", drop = FALSE]
 }
 
+#' @rdname dummyVars
+#' @export
+contr.ltfr <- function (n, contrasts = TRUE, sparse = FALSE)
+{
+  if (is.numeric(n) && length(n) == 1L) {
+    if (n > 1L)
+      levels <- as.character(seq_len(n))
+    else stop("not enough degrees of freedom to define contrasts")
+  }
+  else {
+    levels <- as.character(n)
+    n <- length(n)
+  }
+  contr <- .RDiag(levels, sparse = sparse)
+  if (contrasts) {
+    if (n < 2L) stop(gettextf("contrasts not defined for %d degrees of freedom", n - 1L), domain = NA)
+  }
+  contr
+}
+
+#' @export
+contr.dummy <- function(n, ...)
+{
+  if (is.numeric(n) && length(n) == 1L) {
+    if (n > 1L)
+      levels <- as.character(seq_len(n))
+    else stop("not enough degrees of freedom to define contrasts")
+  }
+  else {
+    levels <- as.character(n)
+    n <- length(n)
+  }
+  out <- diag(n)
+  rownames(out) <- levels
+  colnames(out) <- levels
+  out
+}
+
 

pkg/caret/R/extractPrediction.R

@@ -1,4 +1,4 @@
-
+#' @rdname predict.train
 #' @export
 extractPrediction <- function(models, 
                               testX = NULL, 

pkg/caret/R/extractProb.R

@@ -1,5 +1,5 @@
  ## TODO use foreach to parallelize
-
+#' @rdname predict.train
 #' @importFrom utils flush.console
 #' @export
 extractProb <- function(models, 

pkg/caret/R/filterVarImp.R

@@ -1,8 +1,5 @@
-#' @importFrom ModelMetrics auc
-
 
 #' @importFrom ModelMetrics auc
-
 rocPerCol <- function(dat, cls){
   auc(cls, dat)
 }

pkg/caret/R/gafs.R

@@ -150,6 +150,7 @@ gafs_initial <- function (vars, popSize, ...)  {
   x
 }
 
+#' @rdname gafs_initial
 #' @export
 gafs_lrSelection <-  function (population, fitness,
                                r = NULL,
@@ -169,6 +170,7 @@ gafs_lrSelection <-  function (population, fitness,
   out
 }
 
+#' @rdname gafs_initial
 #' @export
 gafs_spCrossover <- function (population, fitness, parents, ...)  {
   fitness <- fitness[parents]
@@ -193,6 +195,7 @@ gafs_spCrossover <- function (population, fitness, parents, ...)  {
   out
 }
 
+#' @rdname gafs_initial
 #' @export
 gafs_raMutation <- function (population, parent, ...)  {
   mutate <- parent <- as.vector(population[parent, ])
@@ -202,6 +205,7 @@ gafs_raMutation <- function (population, parent, ...)  {
   mutate
 }
 
+#' @rdname gafs_initial
 gafs_nlrSelection <- function (population, fitness, q = 0.25, ...) {
   popSize <- nrow(population)
   rank <- (popSize + 1) - rank(fitness, ties.method = "random")
@@ -213,6 +217,7 @@ gafs_nlrSelection <- function (population, fitness, q = 0.25, ...) {
   return(out)
 }
 
+#' @rdname gafs_initial
 #' @export
 gafs_rwSelection <- function (population, fitness, ...) {
   popSize <- nrow(population)
@@ -224,6 +229,7 @@ gafs_rwSelection <- function (population, fitness, ...) {
   return(out)
 }
 
+#' @rdname gafs_initial
 #' @export
 gafs_tourSelection <- function (population, fitness, k = 3, ...) {
   popSize <- nrow(population)
@@ -237,6 +243,7 @@ gafs_tourSelection <- function (population, fitness, k = 3, ...) {
   return(out)
 }
 
+#' @rdname gafs_initial
 #' @importFrom stats runif
 #' @export
 gafs_uCrossover <- function (population, parents, ...) {
@@ -252,6 +259,7 @@ gafs_uCrossover <- function (population, parents, ...) {
 ###################################################################
 ##
 
+#' @rdname safsControl
 #' @export
 gafsControl <- function(functions = NULL,
                         method = "repeatedcv",
@@ -865,8 +873,7 @@ gafs <- function (x, ...) UseMethod("gafs")
 #' and without each predictor be calculated?
 #' @param gafsControl a list of values that define how this function acts. See
 #' \code{\link{gafsControl}} and URL.
-#' @param list() arguments passed to the classification or regression routine
-#' specified in the function \code{gafsControl$functions$fit}
+#' @param ... additional arguments to be passed to other methods
 #' @return an object of class \code{gafs}
 #' @author Max Kuhn, Luca Scrucca (for GA internals)
 #' @seealso \code{\link{gafsControl}}, \code{\link{predict.gafs}},
@@ -1159,6 +1166,7 @@ gafs <- function (x, ...) UseMethod("gafs")
 #' summary(plot_data)
 #'     }
 #'
+#' @method plot gafs
 #' @export plot.gafs
 plot.gafs <- function(x,
                       metric = x$control$metric["external"],

pkg/caret/R/ggplot.R

@@ -1,3 +1,4 @@
+#' @rdname plot.train
 #' @importFrom stats as.formula
 #' @export
 ggplot.train <- function(data = NULL, mapping = NULL, metric = data$metric[1], plotType = "scatter", output = "layered",
@@ -141,6 +142,7 @@ ggplot.train <- function(data = NULL, mapping = NULL, metric = data$metric[1], p
   out
 }
 
+#' @rdname plot.rfe
 #' @export
 ggplot.rfe <- function(data = NULL, mapping = NULL, metric = data$metric[1],
                        output = "layered", ..., environment = NULL)

pkg/caret/R/icr.R

@@ -85,6 +85,8 @@ icr.formula <- function (formula, data, weights, ...,
     res
 }
 
+#' @rdname icr.formula
+#' @method icr default
 #' @importFrom stats predict lm
 #' @export
 icr.default <- function(x, y, ...)
@@ -132,6 +134,8 @@ print.icr <- function (x, digits = max(3, getOption("digits") - 3), ...)
   invisible(x)
 }
 
+#' @rdname icr.formula
+#' @method predict icr
 #' @importFrom stats .checkMFClasses delete.response model.frame model.matrix predict na.omit fitted
 #' @export
 predict.icr <- function(object, newdata, ...)