Marginal SAGE

SAGE with marginal sampling (features are marginalized independently). This is the standard SAGE implementation.

See also

ConditionalSAGE

Super classes

FeatureImportanceMethod -> SAGE -> MarginalSAGE

Methods

Public methods

• MarginalSAGE$new()

• MarginalSAGE$clone()

Inherited methods

• FeatureImportanceMethod$importance()
• FeatureImportanceMethod$obs_loss()
• FeatureImportanceMethod$print()
• FeatureImportanceMethod$scores()
• SAGE$compute()
• SAGE$convergence()
• SAGE$plot_convergence()
• SAGE$reset()

---

MarginalSAGE$new()

Creates a new instance of the MarginalSAGE class.

Usage

MarginalSAGE$new(
  task,
  learner,
  measure = NULL,
  resampling = NULL,
  features = NULL,
  estimator = c("permutation", "kernel", "exact"),
  n_permutations = NULL,
  n_coalitions = NULL,
  max_features = 12L,
  batch_size = 5000L,
  n_samples = 100L,
  early_stopping = FALSE,
  se_threshold = 0.025,
  min_permutations = 10L,
  check_interval = 1L
)

Arguments

task, learner, measure, resampling, features, estimator, n_permutations, n_coalitions, max_features

Passed to SAGE.

batch_size, n_samples, early_stopping, se_threshold, min_permutations, check_interval

Passed to SAGE.

---

MarginalSAGE$clone()

The objects of this class are cloneable with this method.

Usage

MarginalSAGE$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.

Examples

library(mlr3)
task = tgen("friedman1")$generate(n = 100)
sage = MarginalSAGE$new(
  task = task,
  learner = lrn("regr.ranger", num.trees = 20),
  measure = msr("regr.mse"),
  n_permutations = 2L,
  n_samples = 10
)
#> ℹ No <Resampling> provided, using `resampling = rsmp("holdout", ratio = 2/3)`
#>   (test set size: 33)
sage$compute()