UncertaintyThresholdCriterion<T>

Models & Types · AiDotNet.ActiveLearning.StoppingCriteria

Stopping criterion based on model uncertainty reaching a threshold.

Implements: IUncertaintyBasedCriterion<T>

For Beginners

This criterion stops learning when the model becomes sufficiently confident on the unlabeled data. If the model has low uncertainty on remaining samples, labeling them is unlikely to help much.

How It Works

How It Works:

Key Parameters:

Common Uncertainty Measures:

Constructors

Constructor Summary
UncertaintyThresholdCriterion() Initializes a new UncertaintyThreshold criterion with default parameters.
UncertaintyThresholdCriterion(double uncertaintyThreshold, double requiredConfidentFraction) Initializes a new UncertaintyThreshold criterion with specified parameters.

Properties

Property Summary
T CurrentAverageUncertainty { get; }
T FractionConfident { get; }
T UncertaintyThreshold { get; set; }
string Description { get; }
string Name { get; }

Methods

Method Summary
T GetProgress(ActiveLearningContext<T> context)
bool ShouldStop(ActiveLearningContext<T> context)
void Reset()