mlr (version 2.8)

plotBMRSummary: Plot a benchmark summary.

Description

Creates a scatter plot, where each line refers to a task. On that line the aggregated scores for all learners are plotted, for that task. Optionally, you can apply a rank transformation or just use one of ggplot2's transformations like scale_x_log10.

Usage

plotBMRSummary(bmr, measure = NULL, trafo = "none", order.tsks = NULL, pointsize = 4L, jitter = 0.05)

Arguments

bmr

[BenchmarkResult] Benchmark result.

measure

[Measure] Performance measure. Default is the first measure used in the benchmark experiment.

trafo

[character(1)] Currently either “none” or “rank”, the latter performing a rank transformation (with average handling of ties) of the scores per task. NB: You can add always add scale_x_log10 to the result to put scores on a log scale. Default is “none”.

order.tsks

[character(n.tasks)] Character vector with task.ids in new order.

pointsize

[numeric(1)] Point size for ggplot2 geom_point for data points. Default is 4.

jitter

[numeric(1)] Small vertical jitter to deal with overplotting in case of equal scores. Default is 0.05.

Value

ggplot2 plot object.

See Also

Other benchmark: BenchmarkResult, benchmark, convertBMRToRankMatrix, friedmanPostHocTestBMR, friedmanTestBMR, generateCritDifferencesData, getBMRAggrPerformances, getBMRFeatSelResults, getBMRFilteredFeatures, getBMRLearnerIds, getBMRLearnerShortNames, getBMRLearners, getBMRMeasureIds, getBMRMeasures, getBMRModels, getBMRPerformances, getBMRPredictions, getBMRTaskIds, getBMRTuneResults, plotBMRBoxplots, plotBMRRanksAsBarChart, plotCritDifferences

Other plot: plotBMRBoxplots, plotBMRRanksAsBarChart, plotCalibration, plotCritDifferences, plotFilterValuesGGVIS, plotFilterValues, plotLearningCurveGGVIS, plotLearningCurve, plotPartialPredictionGGVIS, plotPartialPrediction, plotROCCurves, plotThreshVsPerfGGVIS, plotThreshVsPerf

Examples

# see benchmark