Feature Selection Wrapper

Feature selection wrappers can be found in the mlr3fselect packages. The goal is to find the best subset of features with respect to a performance measure in an iterative fashion.

Example Usage

Run a sequential feature selection on the Pima Indian Diabetes data set.

library(mlr3verse)

# retrieve task
task = tsk("pima")

# load learner
learner = lrn("classif.rpart")

# feature selection on the pima indians diabetes data set
instance = fselect(
  method = fs("sequential"),
  task = task,
  learner = learner,
  resampling = rsmp("holdout"),
  measure = msr("classif.ce")
)

# best performing feature subset
instance$result

    age glucose insulin  mass pedigree pregnant pressure triceps    features
1: TRUE    TRUE   FALSE FALSE    FALSE    FALSE    FALSE   FALSE age,glucose
   classif.ce
1:  0.2460938

# subset the task and fit the final model
task$select(instance$result_feature_set)
learner$train(task)

print(learner)

<LearnerClassifRpart:classif.rpart>: Classification Tree
* Model: rpart
* Parameters: xval=0
* Packages: mlr3, rpart
* Predict Types:  [response], prob
* Feature Types: logical, integer, numeric, factor, ordered
* Properties: importance, missings, multiclass, selected_features,
  twoclass, weights