mlr (version 2.19.1)

plotCalibration: Plot calibration data using ggplot2.

Description

Plots calibration data from generateCalibrationData.

Usage

plotCalibration(
  obj,
  smooth = FALSE,
  reference = TRUE,
  rag = TRUE,
  facet.wrap.nrow = NULL,
  facet.wrap.ncol = NULL
)

Value

ggplot2 plot object.

Arguments

obj

(CalibrationData)
Result of generateCalibrationData.

smooth

(logical(1))
Whether to use a loess smoother. Default is FALSE.

reference

(logical(1))
Whether to plot a reference line showing perfect calibration. Default is TRUE.

rag

(logical(1))
Whether to include a rag plot which shows a rug plot on the top which pertains to positive cases and on the bottom which pertains to negative cases. Default is TRUE.

facet.wrap.nrow, facet.wrap.ncol

(integer)
Number of rows and columns for facetting. Default for both is NULL. In this case ggplot's facet_wrap will choose the layout itself.

See Also

Other plot: createSpatialResamplingPlots(), plotBMRBoxplots(), plotBMRRanksAsBarChart(), plotBMRSummary(), plotCritDifferences(), plotLearningCurve(), plotPartialDependence(), plotROCCurves(), plotResiduals(), plotThreshVsPerf()

Other calibration: generateCalibrationData()

Examples

if (FALSE) {
lrns = list(makeLearner("classif.rpart", predict.type = "prob"),
  makeLearner("classif.nnet", predict.type = "prob"))
fit = lapply(lrns, train, task = iris.task)
pred = lapply(fit, predict, task = iris.task)
names(pred) = c("rpart", "nnet")
out = generateCalibrationData(pred, groups = 3)
plotCalibration(out)

fit = lapply(lrns, train, task = sonar.task)
pred = lapply(fit, predict, task = sonar.task)
names(pred) = c("rpart", "lda")
out = generateCalibrationData(pred)
plotCalibration(out)
}