R: Create a bar chart for ranks in a BenchmarkResult.

plotBMRRanksAsBarChart {mlr}

R Documentation

Create a bar chart for ranks in a BenchmarkResult.

Description

Plots a bar chart from the ranks of algorithms. Alternatively, tiles can be plotted for every rank-task combination, see pos for details. In all plot variants the ranks of the learning algorithms are displayed on the x-axis. Areas are always colored according to the learner.id.

Usage

plotBMRRanksAsBarChart(bmr, measure = NULL, ties.method = "average",
  aggregation = "default", pos = "stack", order.lrns = NULL,
  order.tsks = NULL, pretty.names = TRUE)

Arguments

`bmr`	(BenchmarkResult) Benchmark result.
`measure`	(Measure) Performance measure. Default is the first measure used in the benchmark experiment.
`ties.method`	(`character(1)`) See rank for details.
`aggregation`	(`character(1)`) “mean” or “default”. See getBMRAggrPerformances for details on “default”.
`pos`	(`character(1)`) Optionally set how the bars are positioned in ggplot2. Ranks are plotted on the x-axis. “tile” plots a heat map with `task` as the y-axis. Allows identification of the performance in a special task. “stack” plots a stacked bar plot. Allows for comparison of learners within and and across ranks. “dodge” plots a bar plot with bars next to each other instead of stacked bars.
`order.lrns`	(`character(n.learners)`) Character vector with `learner.ids` in new order.
`order.tsks`	(`character(n.tasks)`) Character vector with `task.ids` in new order.
`pretty.names`	(`logical(1)`) Whether to use the short name of the learner instead of its ID in labels. Defaults to `TRUE`.

Value

ggplot2 plot object.

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

Examples

# see benchmark