One Standard Error Rule Callback

Selects the smallest feature set within one standard error of the best as the result. If there are multiple such feature sets with the same number of features, the first one is selected. If the sets have exactly the same performance but different number of features, the one with the smallest number of features is selected.

Source

Kuhn, Max, Johnson, Kjell (2013). “Applied Predictive Modeling.” In chapter Over-Fitting and Model Tuning, 61–92. Springer New York, New York, NY. ISBN 978-1-4614-6849-3.

Examples

clbk("mlr3fselect.one_se_rule")
#> <CallbackBatchFSelect:mlr3fselect.one_se_rule>: One Standard Error Rule Callback
#> * Active Stages: on_optimization_end

# Run feature selection on the pima data set with the callback
instance = fselect(
  fselector = fs("random_search"),
  task = tsk("pima"),
  learner = lrn("classif.rpart"),
  resampling = rsmp ("cv", folds = 3),
  measures = msr("classif.ce"),
  term_evals = 10,
  callbacks = clbk("mlr3fselect.one_se_rule"))
# Smallest feature set within one standard error of the best
instance$result
#>       age glucose insulin   mass pedigree pregnant pressure triceps
#>    <lgcl>  <lgcl>  <lgcl> <lgcl>   <lgcl>   <lgcl>   <lgcl>  <lgcl>
#> 1:   TRUE    TRUE   FALSE   TRUE    FALSE    FALSE    FALSE   FALSE
#>            features n_features classif.ce
#>              <list>     <list>      <num>
#> 1: age,glucose,mass          3  0.2539062