For certain resampling methods, there are default confidence interval methods.
See mlr3::mlr_reflections$default_ci_methods
for a selection.
This measure will select the appropriate CI method depending on the class of the
used Resampling
.
Parameters
Only those from MeasureAbstractCi
.
Super classes
mlr3::Measure
-> mlr3inferr::MeasureAbstractCi
-> Measure
Methods
Method new()
Creates a new instance of this R6 class.
Usage
MeasureCi$new(measure)
Arguments
measure
(
Measure
orcharacter(1)
)
A measure of ID of a measure.
Method aggregate()
Obtain a point estimate, as well as lower and upper CI boundary.
Arguments
rr
(
ResampleResult
)
Resample result.
Examples
rr = resample(tsk("sonar"), lrn("classif.featureless"), rsmp("holdout"))
rr$aggregate(msr("ci", "classif.acc"))
#> classif.acc classif.acc.lower classif.acc.upper
#> 0.3913043 0.2753062 0.5073025
# is the same as:
rr$aggregate(msr("ci.holdout", "classif.acc"))
#> classif.acc classif.acc.lower classif.acc.upper
#> 0.3913043 0.2753062 0.5073025