Filter for Embedded Feature Selection via Variable Importance

Variable Importance filter using embedded feature selection of machine learning algorithms. Takes a mlr3::Learner which is capable of extracting the variable importance (property "importance"), fits the model and extracts the importance values to use as filter scores.

See also

• PipeOpFilter for filter-based feature selection.

• Dictionary of Filters: mlr_filters

Other Filter: Filter, mlr_filters_anova, mlr_filters_auc, mlr_filters_carscore, mlr_filters_carsurvscore, mlr_filters_cmim, mlr_filters_correlation, mlr_filters_disr, mlr_filters_find_correlation, mlr_filters_information_gain, mlr_filters_jmim, mlr_filters_jmi, mlr_filters_kruskal_test, mlr_filters_mim, mlr_filters_mrmr, mlr_filters_njmim, mlr_filters_performance, mlr_filters_permutation, mlr_filters_relief, mlr_filters_selected_features, mlr_filters_variance, mlr_filters

Super class

mlr3filters::Filter -> FilterImportance

Public fields

learner

(mlr3::Learner)
Learner to extract the importance values from.

Methods

Public methods

• FilterImportance$new()

• FilterImportance$clone()

Inherited methods

• mlr3filters::Filter$calculate()
• mlr3filters::Filter$format()
• mlr3filters::Filter$help()
• mlr3filters::Filter$print()

---

Method new()

Create a FilterImportance object.

Usage

FilterImportance$new(learner = mlr3::lrn("classif.featureless"))

Arguments

learner

(mlr3::Learner)
Learner to extract the importance values from.

---

Method clone()

The objects of this class are cloneable with this method.

Usage

FilterImportance$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.

Examples

if (requireNamespace("rpart")) {
  task = mlr3::tsk("iris")
  learner = mlr3::lrn("classif.rpart")
  filter = flt("importance", learner = learner)
  filter$calculate(task)
  as.data.table(filter)
}
#>         feature    score
#> 1:  Petal.Width 88.96940
#> 2: Petal.Length 81.34496
#> 3: Sepal.Length 54.09606
#> 4:  Sepal.Width 36.01309

if (mlr3misc::require_namespaces(c("mlr3pipelines", "rpart", "mlr3learners"), quietly = TRUE)) {
  library("mlr3learners")
  library("mlr3pipelines")
  task = mlr3::tsk("spam")

  learner = mlr3::lrn("classif.rpart")

  # Note: `filter.frac` is selected randomly and should be tuned.

  graph = po("filter", filter = flt("importance", learner = learner), filter.frac = 0.5) %>>%
    po("learner", mlr3::lrn("classif.log_reg"))

  graph$train(task)
}
#> Loading required package: mlr3
#> Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
#> This happened PipeOp classif.log_reg's $train()
#> $classif.log_reg.output
#> NULL
#>