Create Model Explainer

Creates a DALEX-style explainer for any mlr3 learner. The explainer wraps the model for uniform interpretability access.

Usage

create_explainer(learner, data, y, label = NULL, predict_function = NULL)

Arguments

learner

A trained mlr3 Learner or GraphLearner

data

Data frame or matrix used for explanations

y

Response variable (factor for classification)

label

Optional label for the model

predict_function

Custom predict function (optional)

Value

An OmicExplainer object

Examples

if (FALSE) { # \dontrun{
# Create explainer from trained learner
explainer <- create_explainer(
  learner = trained_learner,
  data = train_data,
  y = train_labels
)

# Compute feature importance
importance <- feature_importance(explainer)
} # }