Interface ISelfPacedScheduler<T>

Namespace

AiDotNet.CurriculumLearning.Interfaces

Assembly

AiDotNet.dll

Interface for self-paced curriculum schedulers that adapt based on model performance.

public interface ISelfPacedScheduler<T> : ICurriculumScheduler<T>

Type Parameters

T

The numeric type.

Inherited Members

ICurriculumScheduler<T>.Name

ICurriculumScheduler<T>.CurrentPhase

ICurriculumScheduler<T>.CurrentEpoch

ICurriculumScheduler<T>.TotalPhases

ICurriculumScheduler<T>.CurrentPhaseNumber

ICurriculumScheduler<T>.IsComplete

ICurriculumScheduler<T>.GetDataFraction()

ICurriculumScheduler<T>.GetDifficultyThreshold()

ICurriculumScheduler<T>.StepEpoch(CurriculumEpochMetrics<T>)

ICurriculumScheduler<T>.AdvancePhase()

ICurriculumScheduler<T>.Reset()

ICurriculumScheduler<T>.GetCurrentIndices(int[], int)

ICurriculumScheduler<T>.GetIndicesAtPhase(int[], int, T)

ICurriculumScheduler<T>.GetStatistics()

Properties

GrowthRate

Gets or sets the growth rate for the pace parameter.

T GrowthRate { get; set; }

Property Value

T

PaceParameter

Gets or sets the pace parameter (lambda in self-paced learning).

T PaceParameter { get; set; }

Property Value

T

Remarks

Controls how quickly the curriculum advances. Higher values lead to faster inclusion of harder samples.

Methods

ComputeSampleWeights(Vector<T>)

Computes sample weights for self-paced learning.

Vector<T> ComputeSampleWeights(Vector<T> losses)

Parameters

losses Vector<T>

Per-sample losses.

Returns

Vector<T>

Weights for each sample (0 = ignore, 1 = full weight).