Plots calibration data from generateCalibrationData.
Usage
plotCalibration(
obj,
smooth = FALSE,
reference = TRUE,
rag = TRUE,
facet.wrap.nrow = NULL,
facet.wrap.ncol = NULL
)
Arguments
- obj
(CalibrationData)
Result of generateCalibrationData.- smooth
(
logical(1)
)
Whether to use a loess smoother. Default isFALSE
.- reference
(
logical(1)
)
Whether to plot a reference line showing perfect calibration. Default isTRUE
.- 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 isTRUE
.- facet.wrap.nrow, facet.wrap.ncol
(integer)
Number of rows and columns for facetting. Default for both isNULL
. In this case ggplot'sfacet_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)
}