plotObjectivePairs {TunePareto}

Plot a matrix of Pareto front panels

Description

Plots a matrix of Pareto front panels for each pair of objectives. The plot for n objectives consists of n x n panels, where the panel in row i and column j depicts the Pareto fronts of the i-th and the j-th objective. Each of the panels is drawn in the same way as plotParetoFronts2D.

Usage

plotObjectivePairs(tuneParetoResult, 
                   drawLabels = TRUE, 
                   drawBoundaries = TRUE, 
                   labelPos = 4,
                   fitLabels=TRUE, 
                   cex.conf=0.5, 
                   lty.fronts=1, 
                   pch.fronts=8, 
                   col.fronts,
                   ...)

Arguments

tuneParetoResult

An object of class TuneParetoResult as returned by tunePareto.

drawLabels

If set to true, the descriptions of the configurations are printed next to the points in the plot.

drawBoundaries

If set to true, the upper or lower objective limits supplied in the objectiveBoundaries parameter of tunePareto are drawn as horizontal and vertical lines.

labelPos

The position of the configuration labels in the plot (if drawLabels is true). Values of 1, 2, 3 and 4 denote positions below, to the left of, above and to the right of the points on the Pareto fronts.

fitLabels

If this parameter is true (and drawLabels is true), overlapping or partially hidden configuration labels are removed from the plot to improve the readability of the remaining labels.

cex.conf

The size of the configuration labels in the plots (if drawLabels is true).

lty.fronts

A vector of line types to use for the Pareto fronts. By default, straight lines are drawn for all fronts.

pch.fronts

A vector of symbols to use for points on the Pareto fronts. All points on the same front will have the same symbol. By default, an asterisk is used.

col.fronts

A vector of colors to use for the Pareto fronts. By default, a predefined set of colors is used.

...

Further graphical parameters to be passed to the plot function.

Value

This function does not have a return value.

See Also

tunePareto, plotParetoFronts2D, plotDominationGraph

Examples

# optimize the 'cost' parameter of an SVM according
# to CV error, CV error variance, and CV Specificity 
# on two classes of the 'iris' data set
r <- tunePareto(data = iris[, -ncol(iris)], 
                labels = iris[, ncol(iris)],
                classifier = tunePareto.svm(),
                cost=c(0.001,0.005,0.01,0.05,0.1,0.5,1,5,10,50), 
                objectiveFunctions=list(cvError(10, 10),
                                        cvErrorVariance(10, 10),
                                        cvSpecificity(10, 10, caseClass="virginica")))

# plot the matrix of Pareto fronts
plotObjectivePairs(r)