mlr (version 2.6)

plotBenchmarkSummary: Plot a benchmark-summary.

Description

Plots a BenchmarkSummaryPlot for a selected Measure. Full tiles correspond to the worst performance accross all tasks and learner. The actuall fill corresponds to the proportional performance of the best or worst within the task. Sorting along the x-axis is done accoriding to the algorithms rank.

Usage

plotBenchmarkSummary(obj)

Value

ggplot2 plot object.

References

This plot is analogous to the one described in: Manuel J. A. Eugster, Torsten Hothorn and Friedrich Leisch; Domain-Based Benchmark Experiments:Exploratory and Inferential Analysis, AUSTRIAN JOURNAL OF STATISTICS Volume 41 (2012), Number 1 but does not include any clustering or sorting.

See Also

Other benchmark: BenchmarkResult, benchmark, convertBMRToRankMatrix, friedmanPostHocTestBMR, friedmanTestBMR, generateBenchmarkSummaryData, generateCritDifferencesData, generateRankMatrixAsBarData, getBMRAggrPerformances, getBMRFeatSelResults, getBMRFilteredFeatures, getBMRLearnerIds, getBMRLearners, getBMRMeasureIds, getBMRMeasures, getBMRPerformances, getBMRPredictions, getBMRTaskIds, getBMRTuneResults, plotBenchmarkResult, plotCritDifferences, plotRankMatrixAsBar

Other plot: plotBenchmarkResult, plotCalibration, plotCritDifferences, plotFilterValuesGGVIS, plotFilterValues, plotLearningCurveGGVIS, plotLearningCurve, plotPartialPredictionGGVIS, plotPartialPrediction, plotROCCurves, plotROCRCurvesGGVIS, plotROCRCurves, plotRankMatrixAsBar, plotThreshVsPerfGGVIS, plotThreshVsPerf

Examples

lrns = list(makeLearner("classif.nnet"), makeLearner("classif.rpart"))
tasks = list(iris.task, sonar.task)
rdesc = makeResampleDesc("CV", iters = 2L)
meas = list(acc, mmce, ber, timeboth)
res = benchmark(lrns, tasks, rdesc, meas)
g = generateBenchmarkSummaryData(res, ber, fill = "best")
plotBenchmarkSummary(g)