Extract Inner Tuning Results — extract_inner_tuning

Extract inner tuning results of nested resampling. Implemented for mlr3::ResampleResult and mlr3::BenchmarkResult. The function iterates over the AutoTuner objects and binds the tuning results to a data.table::data.table(). AutoTuner must be initialized with store_tuning_instance = TRUE. For mlr3::BenchmarkResult, the number of the experiment is added to the table.

extract_inner_tuning_results(x)

Arguments

x	(mlr3::ResampleResult \| mlr3::BenchmarkResult) Must contain an AutoTuner.

Value

data.table::data.table().

Examples

task = tsk("iris")
search_space = ParamSet$new(
  params = list(ParamDbl$new("cp", lower = 0.001, upper = 0.1))
)

at = AutoTuner$new(
  learner = lrn("classif.rpart"),
  resampling = rsmp("holdout"),
  measure = msr("classif.ce"),
  terminator = trm("evals", n_evals = 5),
  tuner = tnr("grid_search"),
  search_space = search_space,
  store_tuning_instance = TRUE)

resampling_outer = rsmp("cv", folds = 2)
rr = resample(task, at, resampling_outer, store_models = TRUE)

extract_inner_tuning_results(rr)
#>       cp learner_param_vals  x_domain classif.ce
#> 1: 0.023          <list[2]> <list[1]>       0.04
#> 2: 0.023          <list[2]> <list[1]>       0.08