A simple Dictionary storing objects of class Filter. Each Filter has an associated help page, see mlr_filters_[id].

This dictionary can get populated with additional filters by add-on packages.

For a more convenient way to retrieve and construct filters, see flt().

mlr_filters

Format

R6Class object

Usage

See Dictionary.

See also

Examples

mlr_filters$keys()
#> [1] "anova" "auc" "carscore" "cmim" #> [5] "correlation" "disr" "find_correlation" "importance" #> [9] "information_gain" "jmi" "jmim" "kruskal_test" #> [13] "mim" "mrmr" "njmim" "performance" #> [17] "variance"
as.data.table(mlr_filters)
#> key task_type task_properties param_set #> 1: anova classif <ParamSet> #> 2: auc classif twoclass <ParamSet> #> 3: carscore regr <ParamSet> #> 4: cmim classif,regr <ParamSet> #> 5: correlation regr <ParamSet> #> 6: disr classif <ParamSet> #> 7: find_correlation classif,regr <ParamSet> #> 8: importance classif <ParamSet> #> 9: information_gain classif,regr <ParamSet> #> 10: jmi classif <ParamSet> #> 11: jmim classif <ParamSet> #> 12: kruskal_test classif <ParamSet> #> 13: mim classif <ParamSet> #> 14: mrmr classif <ParamSet> #> 15: njmim classif <ParamSet> #> 16: performance classif <ParamSet> #> 17: variance classif,regr <ParamSet> #> feature_types packages #> 1: integer,numeric stats #> 2: integer,numeric mlr3measures #> 3: numeric care #> 4: integer,numeric,factor,ordered praznik #> 5: integer,numeric stats #> 6: integer,numeric,factor,ordered praznik #> 7: integer,numeric stats #> 8: logical,integer,numeric,factor,ordered rpart #> 9: integer,numeric,factor,ordered FSelectorRcpp #> 10: integer,numeric,factor,ordered praznik #> 11: integer,numeric,factor,ordered praznik #> 12: integer,numeric stats #> 13: integer,numeric,factor,ordered praznik #> 14: integer,numeric,factor,ordered praznik #> 15: integer,numeric,factor,ordered praznik #> 16: logical,integer,numeric,factor,ordered #> 17: integer,numeric stats
mlr_filters$get("mim")
#> <FilterMIM:mim> #> Task Types: classif #> Task Properties: - #> Packages: praznik #> Feature types: integer, numeric, factor, ordered
flt("anova")
#> <FilterAnova:anova> #> Task Types: classif #> Task Properties: - #> Packages: stats #> Feature types: integer, numeric