xplainfi (version 1.0.0)

WVIM: Williamson's Variable Importance Measure (WVIM)

Description

Base class generalizing refit-based variable importance measures. Default corresponds to leaving out each feature n_repeats times, which corresponds to LOCO (Leave One Covariate Out).

Super class

xplainfi::FeatureImportanceMethod -> WVIM

Public fields

direction

(character(1)) Either "leave-out" or "leave-in".

design

(logical()) Feature selection design matrix where TRUE equals "left in" and FALSE "left out". Columns correspond to task$feature_names and the number of rows corresponds to length(features) * n_repeats. The base matrix is created by wvim_design_matrix and then replicated n_repeats times before.

instance

(FSelectInstanceBatchSingleCrit) The mlr3fselect feature selection instance containing also the archive of all evaluations, possible useful for future use. Only stored if store_instance is TRUE.

Methods

Public methods

• WVIM$new()

• WVIM$compute()

• WVIM$clone()

Inherited methods

• xplainfi::FeatureImportanceMethod$importance()
• xplainfi::FeatureImportanceMethod$obs_loss()
• xplainfi::FeatureImportanceMethod$print()
• xplainfi::FeatureImportanceMethod$reset()
• xplainfi::FeatureImportanceMethod$scores()

---

Method new()

Creates a new instance of this R6 class.

Usage

WVIM$new(
  task,
  learner,
  measure = NULL,
  resampling = NULL,
  features = NULL,
  groups = NULL,
  direction = c("leave-out", "leave-in"),
  label = "Williamson's Variable Importance Measure (WVIM)",
  n_repeats = 1L
)

Arguments

task, learner, measure, resampling, features, groups

Passed to FeatureImportanceMethod for construction.

direction

(character(1)) Either "leave-out" or "leave-in".

label

(character(1)) Method label.

n_repeats

(integer(1)) Number of refit iterations per resampling iteration.

---

Method compute()

Computes leave-out or leave-in feature importance. wvim_design_matrix(task$feature_names, "leave-out") corresponds to LOCO.

Usage

WVIM$compute(
  store_models = TRUE,
  store_backends = TRUE,
  store_instance = FALSE
)

Arguments

store_models, store_backends

(logical(1): TRUE) Whether to store fitted models / data backends, passed to mlr3::resample internally backends in resample result. Required for some measures, but may increase memory footprint.

store_instance

(logical(1): FALSE) Whether to store the mlr3fselect::mlr3fselect instance in $instance.

---

Method clone()

The objects of this class are cloneable with this method.

Usage

WVIM$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.