select_features: Select features

Description Usage Arguments Value

Description

Select a subset of model predictor variables (aka features) based on variable importance p-value.

Usage

1
select_features(in_bm, in_lrnid, in_task, pcutoff, inp_resdir = NULL)

Arguments

in_bm

BenchmarkResult containing the ResampleResult for the learner of interest, based on which to select predictor variables. Can also be name of serialized file (.qs) on disk (accessed through qread) containing BenchmarkResult object.

in_lrnid

id of learner to extract from BenchmarkResult (e.g., "oversample.classif.ranger")

in_task

Task containing predictor variables to subset

pcutoff

variable p-value cut-off under which to select variable, see extract_impperf_nestedrf and importance_pvalues for more information on p-value calculation.

inp_resdir

(character) path to where qs file is located (excluding the name of the qs file)

Value

list containing the original in_task and the updated task only included selected predictor variables.


messamat/globalIRmap documentation built on July 4, 2021, 10:48 a.m.