Class UncertaintyCalibrationData<TInput, TOutput>

Namespace

AiDotNet.Models.Inputs

Assembly

AiDotNet.dll

Provides optional calibration data for uncertainty quantification features.

public sealed class UncertaintyCalibrationData<TInput, TOutput>

Type Parameters

TInput

The input type used by the model.

TOutput

The output type used by the model.

Inheritance

object

UncertaintyCalibrationData<TInput, TOutput>

Inherited Members

object.Equals(object)

object.Equals(object, object)

object.GetHashCode()

object.GetType()

object.ReferenceEquals(object, object)

object.ToString()

Remarks

Calibration data is used by certain uncertainty features that require a held-out dataset separate from training data. Examples include conformal prediction and probability calibration.

For Beginners: Think of calibration data as a small "reality check" dataset. The model is trained on training data, then calibration data is used to tune uncertainty-related behavior without overfitting.

Properties

HasLabels

Gets whether classification labels were provided.

public bool HasLabels { get; }

Property Value

bool

HasTargets

Gets whether regression targets were provided.

public bool HasTargets { get; }

Property Value

bool

Labels

Gets the calibration class labels (classification-style calibration).

public Vector<int>? Labels { get; }

Property Value

Vector<int>

Remarks

When provided, the calibration logic treats Y as optional and uses these labels.

X

Gets the calibration inputs.

public TInput X { get; }

Property Value

TInput

Y

Gets the calibration targets (regression-style calibration).

public TOutput Y { get; }

Property Value

TOutput

Methods

ForClassification(TInput, Vector<int>)

Creates calibration data for classification-style calibration (e.g., temperature scaling and conformal prediction sets).

public static UncertaintyCalibrationData<TInput, TOutput> ForClassification(TInput xCalibration, Vector<int> labels)

Parameters

xCalibration TInput

Calibration inputs.

labels Vector<int>

True class labels for calibration samples.

Returns

UncertaintyCalibrationData<TInput, TOutput>

A calibration data instance.

ForRegression(TInput, TOutput)

Creates calibration data for regression-style calibration (e.g., conformal prediction intervals).

public static UncertaintyCalibrationData<TInput, TOutput> ForRegression(TInput xCalibration, TOutput yCalibration)

Parameters

xCalibration TInput

Calibration inputs.

yCalibration TOutput

Calibration targets.

Returns

UncertaintyCalibrationData<TInput, TOutput>

A calibration data instance.