smile.feature.importance
Feature importance and model explanation methods.
Attributes
Members list
Value members
Concrete methods
Creates a KernelSHAP explainer for a scalar function on Array[Double].
Creates a KernelSHAP explainer for a scalar function on Array[Double].
Value parameters
- background
-
the background reference dataset.
- numSamples
-
maximum number of coalitions to evaluate per instance (0 = auto).
- predictor
-
the scalar prediction function.
- ridge
-
L2 regularization parameter for numerical stability.
Attributes
- Returns
-
the KernelSHAP explainer.
Creates a KernelSHAP explainer for a vector-valued function on Array[Double].
Creates a KernelSHAP explainer for a vector-valued function on Array[Double].
Value parameters
- background
-
the background reference dataset.
- numSamples
-
maximum number of coalitions to evaluate per instance (0 = auto).
- predictor
-
the vector prediction function.
- ridge
-
L2 regularization parameter for numerical stability.
Attributes
- Returns
-
the KernelSHAP explainer.