@@ -3,8 +3,8 @@ Type: Package

 Title: A framework for cross-validated classification problems, with

        applications to differential variability and differential

        distribution testing

-Version: 3.1.13

-Date: 2022-08-21

+Version: 3.1.14

+Date: 2022-08-25

 Author: Dario Strbenac, Ellis Patrick, John Ormerod, Graham Mann, Jean Yang

 Maintainer: Dario Strbenac <dario.strbenac@sydney.edu.au>

 VignetteBuilder: knitr

@@ -1,7 +1,7 @@

 #' Plot Receiver Operating Curve Graphs for Classification Results

 #' 

 #' Creates one ROC plot or multiple ROC plots for a list of ClassifyResult

-#' objects.  One plot is created if the data set has two classes and multiple

+#' objects. One plot is created if the data set has two classes and multiple

 #' plots are created if the data set has three or more classes.

 #' 

 #' The scores stored in the results should be higher if the sample is more

@@ -21,13 +21,13 @@

 #' averaged ROC curve.

 #' @param interval Default: 95 (percent). The percent confidence interval to

 #' draw around the averaged ROC curve, if mode is \code{"each"}.

-#' @param comparison The aspect of the experimental design to compare. Can be

+#' @param comparison Default: \code{"auto"}. The aspect of the experimental design to compare. Can be

 #' any characteristic that all results share. If the data set has two classes,

 #' then the slot name with factor levels to be used for colouring the lines.

 #' Otherwise, it specifies the variable used for plot facetting.

-#' @param lineColours A vector of colours for different levels of the

+#' @param lineColours Default: \code{"auto"}. A vector of colours for different levels of the

 #' comparison parameter, or if there are three or more classes, the classes.

-#' If \code{NULL}, a default colour palette is automatically generated.

+#' If \code{"auto"}, a default colour palette is automatically generated.

 #' @param lineWidth A single number controlling the thickness of lines drawn.

 #' @param fontSizes A vector of length 5. The first number is the size of the

 #' title.  The second number is the size of the axes titles and AUC text, if it

@@ -66,7 +66,7 @@

 #'   predicted[c(2, 6), "Healthy"] <- c(0.40, 0.60)

 #'   predicted[c(2, 6), "Cancer"] <- c(0.60, 0.40)

 #'   result2 <- ClassifyResult(DataFrame(characteristic = c("Data Set", "Selection Name", "Classifier Name", "Cross-validation"),

-#'                                       value = c("Example", "Bartlett Test", "Differential Variability", "2-fold")),

+#'                                       value = c("Melanoma", "Bartlett Test", "Differential Variability", "2-fold")),

 #'                             LETTERS[1:20], paste("Gene", LETTERS[1:10]), list(paste("Gene", LETTERS[1:10]), paste("Gene", LETTERS[c(5:1, 6:10)])),

 #'                             list(paste("Gene", LETTERS[1:3]), paste("Gene", LETTERS[1:5])),

 #'                             list(function(oracle){}), NULL, predicted, actual)

@@ -80,7 +80,7 @@ setGeneric("ROCplot", function(results, ...) standardGeneric("ROCplot"))

 #' @export

 setMethod("ROCplot", "list", 

           function(results, mode = c("merge", "average"), interval = 95,

-                   comparison = "Classifier Name", lineColours = NULL,

+                   comparison = "auto", lineColours = "auto",

                    lineWidth = 1, fontSizes = c(24, 16, 12, 12, 12), labelPositions = seq(0.0, 1.0, 0.2),

                    plotTitle = "ROC", legendTitle = NULL, xLabel = "False Positive Rate", yLabel = "True Positive Rate", showAUC = TRUE)

 {

@@ -89,7 +89,15 @@ setMethod("ROCplot", "list",

   if(!requireNamespace("scales", quietly = TRUE))

     stop("The package 'scales' could not be found. Please install it.")

   mode <- match.arg(mode)

-                      

+  characteristicsCounts <- table(unlist(lapply(results, function(result) result@characteristics[["characteristic"]])))

+  if(comparison == "auto")

+  {

+    if(max(characteristicsCounts) == length(results))

+      comparison <- names(characteristicsCounts)[characteristicsCounts == max(characteristicsCounts)][1]

+    else

+      stop("No characteristic is present for all results but must be.")

+  }

+               

   ggplot2::theme_set(ggplot2::theme_classic() + ggplot2::theme(panel.border = ggplot2::element_rect(fill = NA)))

   distinctClasses <- levels(actualOutcome(results[[1]]))

   numberDistinctClasses <- length(distinctClasses)

@@ -181,8 +189,8 @@ setMethod("ROCplot", "list",

   else

     lineColour <- comparison

-  if(is.null(lineColours))

-      lineColours <- scales::hue_pal()(ifelse(lineColour == "class", numberDistinctClasses, length(unique(comparisonValues))))

+  if(lineColours == "auto")

+      lineColours <- scales::hue_pal()(ifelse(lineColour == "class", numberDistinctClasses, max(characteristicsCounts)))

   if(is.null(legendTitle))

     legendTitle <- ifelse(lineColour == "class", "Class", comparisonName)

@@ -207,8 +215,6 @@ setMethod("ROCplot", "list",

   comparison <- rlang::sym(comparison)

   ROCplots <- lapply(plotDataSets, function(plotData)

               {

-

-

                 ROCplot <- ggplot2::ggplot(plotData, ggplot2::aes(x = FPR, y = TPR, colour = !!lineColour)) +

                            ggplot2::geom_line(size = lineWidth) + ggplot2::xlab(NULL) + ggplot2::ylab(NULL) + ggplot2::labs(colour = legendTitle) + ggplot2::geom_segment(x = 0, y = 0, xend = 1, yend = 1, size = lineWidth, colour = "black") + ggplot2::scale_x_continuous(breaks = labelPositions, limits = c(0, 1)) +  ggplot2::scale_y_continuous(breaks = labelPositions, limits = c(0, 1)) +

                            ggplot2::theme(axis.text = ggplot2::element_text(colour = "black", size = fontSizes[3]), legend.position = c(1, 0), legend.justification = c(1, 0), legend.background = ggplot2::element_rect(fill = "transparent"), legend.title = ggplot2::element_text(size = fontSizes[4], hjust = 0), legend.text = ggplot2::element_text(size = fontSizes[5])) + ggplot2::guides(colour = ggplot2::guide_legend(title.hjust = 0.5)) + ggplot2::scale_colour_manual(values = lineColours)

@@ -641,13 +641,15 @@ setClass("TrainParams", representation(

 #' @docType class

 #' @section Constructor:

 #' \describe{

-#' \item{}{\preformatted{TrainParams(classifier, characteristics = DataFrame(),

-#' intermediate = character(0), getFeatures = NULL, ...)}

+#' \item{}{\preformatted{TrainParams(classifier, balancing = c("downsample", "upsample", "none"), characteristics = DataFrame(),

+#' intermediate = character(0), tuneParams = NULL, getFeatures = NULL, ...)}

 #' Creates a \code{TrainParams} object which stores the function which will do the

 #' classifier building and parameters that the function will use.

 #' \describe{

 #' \item{\code{classifier}}{A character keyword referring to a registered classifier. See \code{\link{available}}

 #' for valid keywords.}

+#' \item{\code{balancing}}{Default: \code{"downsample"}. A keyword specifying how to handle class imbalance for data sets with categorical outcome.

+#' Valid values are \code{"downsample"}, \code{"upsample"} and \code{"none"}.}

 #' \item{\code{characteristics}}{A \code{\link{DataFrame}} describing the

 #' characteristics of the classifier used. First column must be named \code{"charateristic"}

 #' and second column must be named \code{"value"}. If using wrapper functions for classifiers

@@ -656,6 +658,9 @@ setClass("TrainParams", representation(

 #' \item{\code{intermediate}}{Character vector. Names of any variables created

 #' in prior stages by \code{\link{runTest}} that need to be passed to

 #' \code{classifier}.}

+#' \item{\code{tuneParams}}{A list specifying tuning parameters required during feature selection. The names of

+#' the list are the names of the parameters and the vectors are the values of the parameters to try. All possible

+#' combinations are generated.}

 #' \item{\code{getFeatures}}{A function may be specified that extracts the selected

 #' features from the trained model. This is relevant if using a classifier that does

 #' feature selection within training (e.g. random forest). The function must return a

@@ -964,7 +969,7 @@ setClassUnion("ModellingParamsOrNULL", c("ModellingParams", "NULL"))

 #'   #if(require(sparsediscrim))

 #'   #{

 #'     data(asthma)

-#'     classified <- crossValidate(measurements, classes)

+#'     classified <- crossValidate(measurements, classes, nRepeats = 5)

 #'     class(classified)

 #'   #}

 #'   

@@ -76,6 +76,7 @@

     "GLM", "Logistic regression.",

     "elasticNetGLM", "Elastic net GLM multinomial regression.",

     "SVM", "Support Vector Machine.",

+    "NSC", "Nearest Shrunken Centroids.",

     "naiveBayes", "Naive Bayes kernel feature voting classifier.",

     "mixturesNormals", "Mixture of normals feature voting classifier.",

     "CoxPH", "Cox proportional hazards.",

@@ -36,6 +36,6 @@ randomForestPredictInterface <- function(forest, measurementsTest, ..., returnTy

 forestFeatures <- function(forest)

                   {

                     rankedFeaturesIndices <- order(randomForest::importance(forest), decreasing = TRUE)

-                    selectedFeaturesIndices <- which(randomForest::varUsed(forest) > 0)

+                    selectedFeaturesIndices <- randomForest::varUsed(forest, count = FALSE)

                     list(rankedFeaturesIndices, selectedFeaturesIndices)

                   }

\ No newline at end of file

@@ -15,7 +15,7 @@

 #' \code{characteristicsList['x']} to aggregate to a single number by taking

 #' the mean. This is particularly meaningful when the cross-validation is

 #' leave-k-out, when k is small.

-#' @param performanceName Default: "auto". The name of the

+#' @param performanceName Default: \code{"auto"}. The name of the

 #' performance measure or "auto". If the results are classification then

 #' balanced accuracy will be displayed. Otherwise, the results would be survival risk

 #' predictions and then C-index will be displayed. This is one of the names printed

@@ -70,11 +70,11 @@

 #' \code{FALSE}, the x-axis labels are hidden.

 #' @param showYtickLabels Logical. Default: \code{TRUE}. If set to

 #' \code{FALSE}, the y-axis labels are hidden.

-#' @param xLabelPositions Either "auto" or a vector of values. The positions of

-#' labels on the x-axis.  If "auto", the placement of labels is automatically

+#' @param xLabelPositions Either \code{"auto"} or a vector of values. The positions of

+#' labels on the x-axis.  If \code{"auto"}, the placement of labels is automatically

 #' calculated.

-#' @param yLabelPositions Either "auto" or a vector of values. The positions of

-#' labels on the y-axis.  If "auto", the placement of labels is automatically

+#' @param yLabelPositions Either \code{"auto"} or a vector of values. The positions of

+#' labels on the y-axis.  If \code{"auto"}, the placement of labels is automatically

 #' calculated.

 #' @param fontSizes A vector of length 5. The first number is the size of the

 #' title.  The second number is the size of the axes titles. The third number

@@ -167,7 +167,7 @@ input data. Autmomatically reducing to smaller number.")

       modellingParams@selectParams <- .addIntermediates(modellingParams@selectParams)

     topFeatures <- tryCatch(.doSelection(measurementsTrain, outcomeTrain, crossValParams, modellingParams, verbose),

-                            error = function(error) error[["message"]]) 

+                            error = function(error) error[["message"]])

     if(is.character(topFeatures)) return(topFeatures) # An error occurred.

     rankedFeaturesIndices <- topFeatures[[1]] # Extract for result object.

@@ -186,10 +186,11 @@ input data. Autmomatically reducing to smaller number.")

     modellingParams@trainParams <- .addIntermediates(modellingParams@trainParams)

   if(!is.null(tuneDetailsSelect))

   {

-    avoidTune <- match(colnames(tuneDetailsSelect), names(modellingParams@trainParams@tuneParams))

+    tuneDetailsSelectUse <- tuneDetailsSelect[["tuneCombinations"]][tuneDetailsSelect[["bestIndex"]], , drop = FALSE]

+    avoidTune <- match(colnames(tuneDetailsSelectUse), names(modellingParams@trainParams@tuneParams))

     if(any(!is.na(avoidTune)))

     {

-      modellingParams@trainParams@otherParams <- c(modellingParams@trainParams@otherParams, tuneDetailsSelect[!is.na(avoidTune)])

+      modellingParams@trainParams@otherParams <- c(modellingParams@trainParams@otherParams, tuneDetailsSelectUse[!is.na(avoidTune)])

       modellingParams@trainParams@tuneParams <- modellingParams@trainParams@tuneParams[-na.omit(avoidTune)]

       if(length(modellingParams@trainParams@tuneParams) == 0) modellingParams@trainParams@tuneParams <- NULL

     }

@@ -209,7 +210,7 @@ input data. Autmomatically reducing to smaller number.")

     if(length(extras) > 0)

       extrasList <- mget(setdiff(names(extras), "..."))

-    rankedChosenList <- do.call(modellingParams@trainParams@getFeatures, c(trained[1], extrasList))

+    rankedChosenList <- do.call(modellingParams@trainParams@getFeatures, c(unname(trained[1]), extrasList))

     rankedFeaturesIndices <- rankedChosenList[[1]]

     selectedFeaturesIndices <- rankedChosenList[[2]]

   }

@@ -304,8 +305,8 @@ input data. Autmomatically reducing to smaller number.")

   } else { # runTest executed by the end user. Create a ClassifyResult object.

     # Only one training, so only one tuning choice, which can be summarised in characteristics.

     modParamsList <- list(modellingParams@transformParams, modellingParams@selectParams, modellingParams@trainParams, modellingParams@predictParams)

-    if(!is.null(tuneDetails)) characteristics <- rbind(characteristics, data.frame(characteristic = colnames(tuneDetails),

-                                                                                   value = unlist(tuneDetails)))

+    if(!is.null(tuneDetails)) characteristics <- rbind(characteristics, data.frame(characteristic = colnames(tuneDetails[["tuneCombinations"]]),

+                                                                                   value = unlist(tuneDetails[["tuneCombinations"]][tuneDetails[["bestIndex"]], ])))

     autoCharacteristics <- do.call(rbind, lapply(modParamsList, function(stageParams) if(!is.null(stageParams) && !is(stageParams, "PredictParams")) stageParams@characteristics))

     characteristics <- .filterCharacteristics(characteristics, autoCharacteristics)

     characteristics <- rbind(characteristics, S4Vectors::DataFrame(characteristic = "Cross-validation", value = "Independent Set"))

@@ -1,6 +1,7 @@

 # Random Forest

 RFparams <- function() {

-    trainParams <- TrainParams(randomForestTrainInterface, tuneParams = list(mTryProportion = c(0.25, 0.33, 0.50, 0.66, 0.75, 1.00), ntree = seq(100, 500, 100)))

+    trainParams <- TrainParams(randomForestTrainInterface, tuneParams = list(mTryProportion = c(0.25, 0.33, 0.50, 0.66, 0.75, 1.00), ntree = seq(100, 500, 100)),

+                               getFeatures = forestFeatures)

     predictParams <- PredictParams(randomForestPredictInterface)

     return(list(trainParams = trainParams, predictParams = predictParams))

@@ -31,7 +32,7 @@ GLMparams <- function() {

 # Elastic net GLM

 elasticNetGLMparams <- function() {

-    trainParams <- TrainParams(elasticNetGLMtrainInterface)

+    trainParams <- TrainParams(elasticNetGLMtrainInterface, getFeatures = elasticNetFeatures)

     predictParams <- PredictParams(elasticNetGLMpredictInterface)

     return(list(trainParams = trainParams, predictParams = predictParams))

@@ -45,6 +46,14 @@ SVMparams = function() {

     return(list(trainParams = trainParams, predictParams = predictParams))

 }

+# Nearest Shrunken Centroid

+NSCparams = function() {

+    trainParams <- TrainParams(NSCtrainInterface, getFeatures = NSCfeatures)

+    predictParams <- PredictParams(NSCpredictInterface)

+    

+    return(list(trainParams = trainParams, predictParams = predictParams))

+}

+

 # Diagonal Linear Discriminant Analysis

 DLDAparams = function() {

     trainParams <- TrainParams(DLDAtrainInterface)

@@ -193,14 +193,14 @@

     if(attr(featureRanking, "name") == "previousSelection") # Actually selection not ranking.

       return(list(NULL, rankings[[1]], NULL))

-    if(tuneMode == "none") # Actually selection not ranking.

+    if(tuneMode == "none") # No parameters to choose between.

         return(list(NULL, rankings[[1]], NULL))

     tuneParamsTrain <- list(topN = topNfeatures)

     tuneParamsTrain <- append(tuneParamsTrain, modellingParams@trainParams@tuneParams)

     tuneCombosTrain <- expand.grid(tuneParamsTrain, stringsAsFactors = FALSE)  

     modellingParams@trainParams@tuneParams <- NULL

-    bestPerformers <- sapply(rankings, function(rankingsVariety)

+    allPerformanceTables <- lapply(rankings, function(rankingsVariety)

     {

       # Creates a matrix. Columns are top n features, rows are varieties (one row if None).

       performances <- sapply(1:nrow(tuneCombosTrain), function(rowIndex)

@@ -240,20 +240,22 @@

        })

         bestOne <- ifelse(betterValues == "lower", which.min(performances)[1], which.max(performances)[1])

-        c(bestOne, performances[bestOne])

+        list(data.frame(tuneCombosTrain, performance = performances), bestOne)

       })

-      tunePick <- ifelse(betterValues == "lower", which.min(bestPerformers[2, ])[1], which.max(bestPerformers[2, ])[1])

+      tablesBestMetrics <- sapply(allPerformanceTables, function(tableIndexPair) tableIndexPair[[1]][tableIndexPair[[2]], "performance"])

+      tunePick <- ifelse(betterValues == "lower", which.min(tablesBestMetrics)[1], which.max(tablesBestMetrics)[1])

       if(verbose == 3)

          message("Features selected.")

-      tuneRow <- tuneCombosTrain[bestPerformers[1, tunePick], , drop  = FALSE]

-      if(ncol(tuneRow) > 1) tuneDetails <- tuneRow[, -1, drop = FALSE] else tuneDetails <- NULL

+      tuneDetails <- allPerformanceTables[[tunePick]] # List of length 2.

       rankingUse <- rankings[[tunePick]]

-      selectionIndices <- rankingUse[1:tuneRow[, "topN"]]

+      selectionIndices <- rankingUse[1:(tuneDetails[[1]][tuneDetails[[2]], "topN"])]

+      names(tuneDetails) <- c("tuneCombinations", "bestIndex")

+      colnames(tuneDetails[[1]])[ncol(tuneDetails[[1]])] <- performanceType

       list(ranked = rankingUse, selected = selectionIndices, tune = tuneDetails)

     } else if(is.list(featureRanking)) { # It is a list of functions for ensemble selection.

       featuresIndiciesLists <- mapply(function(selector, selParams)

@@ -296,8 +298,7 @@

       list(NULL, selectionIndices, NULL)

     } else { # Previous selection

-      selectedFeatures <- 

-      list(NULL, selectionIndices, NULL)

+      selectedFeatures <- list(NULL, selectionIndices, NULL)

     }

 }

@@ -315,7 +316,7 @@

 # within the same function, so test samples are also passed in case they are needed.

 .doTrain <- function(measurementsTrain, outcomeTrain, measurementsTest, outcomeTest, modellingParams, verbose)

 {

-  tuneChosen <- NULL

+  tuneDetails <- NULL

   if(!is.null(modellingParams@trainParams@tuneParams) && is.null(modellingParams@selectParams))

   {

     performanceType <- modellingParams@trainParams@tuneParams[["performanceType"]]

@@ -346,9 +347,14 @@

         median(performances(result)[[performanceType]])

       }

     })

+    allPerformanceTable <- data.frame(tuneCombos, performances)

+    colnames(allPerformanceTable)[ncol(allPerformanceTable)] <- performanceType

+    

     betterValues <- .ClassifyRenvir[["performanceInfoTable"]][.ClassifyRenvir[["performanceInfoTable"]][, "type"] == performanceType, "better"]

     bestOne <- ifelse(betterValues == "lower", which.min(performances)[1], which.max(performances)[1])

     tuneChosen <- tuneCombos[bestOne, , drop = FALSE]

+    tuneDetails <- list(tuneCombos, bestOne)

+    names(tuneDetails) <- c("tuneCombinations", "bestIndex")

     modellingParams@trainParams@otherParams <- tuneChosen

   }

@@ -367,7 +373,7 @@

   if(verbose >= 2)

     message("Training completed.")  

-  list(model = trained, tune = tuneChosen)

+  list(model = trained, tune = tuneDetails)

 }

 # Creates a function call to a prediction function.

@@ -566,6 +572,7 @@

         "GLM" = GLMparams(),

         "elasticNetGLM" = elasticNetGLMparams(),

         "SVM" = SVMparams(),

+        "NSC" = NSCparams(),

         "DLDA" = DLDAparams(),

         "naiveBayes" = naiveBayesParams(),

         "mixturesNormals" = mixModelsParams(),

@@ -104,7 +104,7 @@ of predictions made during the cross-validation procedure.}}

   #if(require(sparsediscrim))

   #{

     data(asthma)

-    classified <- crossValidate(measurements, classes)

+    classified <- crossValidate(measurements, classes, nRepeats = 5)

     class(classified)

   #}

@@ -9,8 +9,8 @@

   results,

   mode = c("merge", "average"),

   interval = 95,

-  comparison = "Classifier Name",

-  lineColours = NULL,

+  comparison = "auto",

+  lineColours = "auto",

   lineWidth = 1,

   fontSizes = c(24, 16, 12, 12, 12),

   labelPositions = seq(0, 1, 0.2),

@@ -33,14 +33,14 @@ averaged ROC curve.}

 \item{interval}{Default: 95 (percent). The percent confidence interval to

 draw around the averaged ROC curve, if mode is \code{"each"}.}

-\item{comparison}{The aspect of the experimental design to compare. Can be

+\item{comparison}{Default: \code{"auto"}. The aspect of the experimental design to compare. Can be

 any characteristic that all results share. If the data set has two classes,

 then the slot name with factor levels to be used for colouring the lines.

 Otherwise, it specifies the variable used for plot facetting.}

-\item{lineColours}{A vector of colours for different levels of the

+\item{lineColours}{Default: \code{"auto"}. A vector of colours for different levels of the

 comparison parameter, or if there are three or more classes, the classes.

-If \code{NULL}, a default colour palette is automatically generated.}

+If \code{"auto"}, a default colour palette is automatically generated.}

 \item{lineWidth}{A single number controlling the thickness of lines drawn.}

@@ -71,7 +71,7 @@ device, if \code{plot} is \code{TRUE}.

 }

 \description{

 Creates one ROC plot or multiple ROC plots for a list of ClassifyResult

-objects.  One plot is created if the data set has two classes and multiple

+objects. One plot is created if the data set has two classes and multiple

 plots are created if the data set has three or more classes.

 }

 \details{

@@ -103,7 +103,7 @@ considered simultaneously, to calculate one curve per classification.

   predicted[c(2, 6), "Healthy"] <- c(0.40, 0.60)

   predicted[c(2, 6), "Cancer"] <- c(0.60, 0.40)

   result2 <- ClassifyResult(DataFrame(characteristic = c("Data Set", "Selection Name", "Classifier Name", "Cross-validation"),

-                                      value = c("Example", "Bartlett Test", "Differential Variability", "2-fold")),

+                                      value = c("Melanoma", "Bartlett Test", "Differential Variability", "2-fold")),

                             LETTERS[1:20], paste("Gene", LETTERS[1:10]), list(paste("Gene", LETTERS[1:10]), paste("Gene", LETTERS[c(5:1, 6:10)])),

                             list(paste("Gene", LETTERS[1:3]), paste("Gene", LETTERS[1:5])),

                             list(function(oracle){}), NULL, predicted, actual)

@@ -15,13 +15,15 @@ The empty constructor is provided for convenience.

 \section{Constructor}{

 \describe{

-\item{}{\preformatted{TrainParams(classifier, characteristics = DataFrame(),

-intermediate = character(0), getFeatures = NULL, ...)}

+\item{}{\preformatted{TrainParams(classifier, balancing = c("downsample", "upsample", "none"), characteristics = DataFrame(),

+intermediate = character(0), tuneParams = NULL, getFeatures = NULL, ...)}

 Creates a \code{TrainParams} object which stores the function which will do the

 classifier building and parameters that the function will use.

 \describe{

 \item{\code{classifier}}{A character keyword referring to a registered classifier. See \code{\link{available}}

 for valid keywords.}

+\item{\code{balancing}}{Default: \code{"downsample"}. A keyword specifying how to handle class imbalance for data sets with categorical outcome.

+Valid values are \code{"downsample"}, \code{"upsample"} and \code{"none"}.}

 \item{\code{characteristics}}{A \code{\link{DataFrame}} describing the

 characteristics of the classifier used. First column must be named \code{"charateristic"}

 and second column must be named \code{"value"}. If using wrapper functions for classifiers

@@ -30,6 +32,9 @@ therefore it is not necessary to specify it.}

 \item{\code{intermediate}}{Character vector. Names of any variables created

 in prior stages by \code{\link{runTest}} that need to be passed to

 \code{classifier}.}

+\item{\code{tuneParams}}{A list specifying tuning parameters required during feature selection. The names of

+the list are the names of the parameters and the vectors are the values of the parameters to try. All possible

+combinations are generated.}

 \item{\code{getFeatures}}{A function may be specified that extracts the selected

 features from the trained model. This is relevant if using a classifier that does

 feature selection within training (e.g. random forest). The function must return a

@@ -29,7 +29,7 @@

 \item{...}{Not used by end user.}

-\item{performanceName}{Default: "auto". The name of the

+\item{performanceName}{Default: \code{"auto"}. The name of the

 performance measure or "auto". If the results are classification then

 balanced accuracy will be displayed. Otherwise, the results would be survival risk

 predictions and then C-index will be displayed. This is one of the names printed

@@ -114,12 +114,12 @@ plot. Ignored if the feature's measurements are categorical.}

 \item{showYtickLabels}{Logical. Default: \code{TRUE}. If set to

 \code{FALSE}, the y-axis labels are hidden.}

-\item{xLabelPositions}{Either "auto" or a vector of values. The positions of

-labels on the x-axis.  If "auto", the placement of labels is automatically

+\item{xLabelPositions}{Either \code{"auto"} or a vector of values. The positions of

+labels on the x-axis.  If \code{"auto"}, the placement of labels is automatically

 calculated.}

-\item{yLabelPositions}{Either "auto" or a vector of values. The positions of

-labels on the y-axis.  If "auto", the placement of labels is automatically

+\item{yLabelPositions}{Either \code{"auto"} or a vector of values. The positions of

+labels on the y-axis.  If \code{"auto"}, the placement of labels is automatically

 calculated.}

 \item{fontSizes}{A vector of length 5. The first number is the size of the

@@ -15,8 +15,8 @@

   referenceLevel = NULL,

   characteristicsList = list(),

   orderingList = list(),

-  sizesList = list(lineWidth = 1, pointSize = 2, legendLinesPointsSize = 1, fonts = c(24,

-    16, 12, 12, 12, 16)),

+  sizesList = list(lineWidth = 1, pointSize = 2, legendLinesPointsSize = 1, fonts =

+    c(24, 16, 12, 12, 12, 16)),

   lineColours = NULL,

   xLabelPositions = seq(10, 100, 10),

   yMax = 100,