LocalTCAVExplanation<T>
Models & Types · AiDotNet.Interpretability.Explainers
Represents a local TCAV explanation for a single input.
Constructors
| Constructor |
Summary |
LocalTCAVExplanation(Vector<T> input, string conceptName, int targetClass, T directionalDerivative, T conceptProjection, T prediction, string influenceDirection) |
Initializes a new local TCAV explanation. |
Properties
| Property |
Summary |
T ConceptProjection { get; } |
Gets the projection of activations onto the concept direction. |
T DirectionalDerivative { get; } |
Gets the directional derivative (concept sensitivity). |
T Prediction { get; } |
Gets the prediction score for the target class. |
Vector<T> Input { get; } |
Gets the input that was explained. |
int TargetClass { get; } |
Gets the target class. |
string ConceptName { get; } |
Gets the concept name. |
string InfluenceDirection { get; } |
Gets the influence direction (“positive” or “negative”). |
Methods
| Method |
Summary |
string ToString() |
Returns a human-readable summary. |