SelfInfluenceResult<T>

Models & Types · AiDotNet.Interpretability.Explainers

Result of self-influence computation.

Constructors

Constructor Summary
SelfInfluenceResult(Vector<T> selfInfluences, Matrix<T> trainingData, Vector<T> trainingLabels) Initializes a new self-influence result.

Properties

Property Summary
Matrix<T> TrainingData { get; } Gets the training data.
Vector<T> SelfInfluences { get; } Gets the self-influence score for each training sample.
Vector<T> TrainingLabels { get; } Gets the training labels.

Methods

Method Summary
IEnumerable<ValueTuple<int, T, Vector<T>, T>> GetPotentiallyProblematic(int k) Gets samples most likely to be mislabeled or problematic.
string ToString() Returns a human-readable summary.