Calculates the Correlation-Adjusted (marginal) coRelation scores (short CAR scores) implemented in care::carscore() in package care. The CAR scores for a set of features are defined as the correlations between the target and the decorrelated features. The filter returns the absolute value of the calculated scores.

Argument verbose defaults to FALSE.

Format

R6::R6Class inheriting from Filter.

Construction

FilterCarScore$new()
mlr_filters$get("carscore")
flt("carscore")

See also

Examples

task = mlr3::tsk("mtcars") filter = flt("carscore") filter$calculate(task) head(as.data.table(filter), 3)
#> feature score #> 1: wt 0.4144012 #> 2: hp 0.3174307 #> 3: cyl 0.3102745
## changing filter settings filter = flt("carscore") filter$param_set$values = list("diagonal" = TRUE) filter$calculate(task)
#> Estimating optimal shrinkage intensity lambda (correlation matrix): 0.0707
head(as.data.table(filter), 3)
#> feature score #> 1: wt 0.8062818 #> 2: cyl 0.7918806 #> 3: disp 0.7875962