Details
Calculates Leave-One-Covariate-Out (LOCO) scores.
Methods
Inherited methods
Method new()
Creates a new instance of this R6 class.
Usage
LOCO$new(task, learner, measure, resampling = NULL, features = NULL)
Arguments
task, learner, measure, resampling, features
Passed to FeatureImportanceLearner
for construction.
Method compute()
A short description...
Usage
LOCO$compute(relation = c("difference", "ratio"))
Arguments
relation
(character(1)) Calculate "difference"
(default) or "ratio"
of
original scores and scores after permutation
Method clone()
The objects of this class are cloneable with this method.
Arguments
deep
Whether to make a deep clone.
Examples
library(mlr3)
loco = LOCO$new(
task = tsk("zoo"),
learner = lrn("classif.rpart"),
measure = msr("classif.ce")
)
loco$compute()
#> airborne aquatic backbone breathes catsize domestic eggs feathers
#> 0 0 0 0 0 0 0 0
#> fins hair legs milk predator tail toothed venomous
#> 0 0 0 0 0 0 0 0