DiversitySampling<T>

Models & Types · AiDotNet.ActiveLearning

Implements Diversity Sampling for active learning.

Implements: IActiveLearningStrategy<T>

For Beginners

Diversity Sampling selects samples that are representative of different regions in the input space. Instead of focusing on uncertain samples near the decision boundary, diversity sampling ensures good coverage of the data distribution.

How It Works

How it works:

Strategies:

Reference: Sener and Savarese, “Active Learning for Convolutional Neural Networks: A Core-Set Approach” (2018). ICLR.

Constructors

Constructor Summary
DiversitySampling(DiversityMethod<T> method, DistanceMetric<T> distanceMetric) Initializes a new instance of the DiversitySampling class.

Properties

Property Summary
T CoverageRadius { get; } Gets the coverage radius from the last selection.
bool UseBatchDiversity { get; set; }
string Name { get; }

Methods

Method Summary
Dictionary<string, T> GetSelectionStatistics()
Vector<T> ComputeInformativenessScores(IFullModel<T, Tensor<T>, Tensor<T>> model, Tensor<T> unlabeledPool)
int[] SelectSamples(IFullModel<T, Tensor<T>, Tensor<T>> model, Tensor<T> unlabeledPool, int batchSize)