Samplers

Helpers & Utilities · AiDotNet.Data.Sampling

Static factory class for creating data samplers with beginner-friendly methods.

For Beginners

Sampling strategies determine how you pick data from your dataset. Different strategies can help with:

  • Class imbalance (use weighted sampling)
  • Curriculum learning (start with easy examples, progress to hard)
  • Active learning (focus on uncertain examples)

Common Patterns:

// Random sampling (default, good for most cases)
var dataSize = 1000;
var randomSampler = Samplers.Random(dataSize);

// Balanced sampling for imbalanced classes
var labels = new[] { 0, 1, 0, 2, 1, 0 };
var balancedSampler = Samplers.Balanced(labels, numClasses: 3);

// Curriculum learning (easy to hard)
var difficulties = new[] { 0.1, 0.4, 0.2, 0.9, 0.6, 0.3 };
var curriculumSampler = Samplers.Curriculum(difficulties, totalEpochs: 100);

How It Works

Samplers provides factory methods for creating various sampling strategies used during training. These samplers control which data points are selected and in what order.

Methods

Method Summary
ActiveLearningSampler<T> ActiveLearning(int datasetSize, ActiveLearningStrategy strategy, double diversityWeight, int? seed) Creates an active learning sampler that prioritizes uncertain samples.
ActiveLearningSampler<double> ActiveLearning(int datasetSize, ActiveLearningStrategy strategy, double diversityWeight, int? seed) Creates an active learning sampler that prioritizes uncertain samples.
CurriculumSampler<T> Curriculum(IEnumerable<T> difficulties, int totalEpochs, CurriculumStrategy strategy, int? seed) Creates a curriculum learning sampler that starts with easy samples.
CurriculumSampler<double> Curriculum(IEnumerable<double> difficulties, int totalEpochs, CurriculumStrategy strategy, int? seed) Creates a curriculum learning sampler that starts with easy samples.
ImportanceSampler<T> Importance(int datasetSize, double smoothingFactor, bool stabilize, int? seed) Creates an importance sampler that prioritizes high-loss samples.
ImportanceSampler<double> Importance(int datasetSize, double smoothingFactor, bool stabilize, int? seed) Creates an importance sampler that prioritizes high-loss samples.
RandomSampler Random(int dataSize, int? seed) Creates a random sampler that shuffles data each epoch.
SelfPacedSampler<T> SelfPaced(int datasetSize, T initialLambda, T lambdaGrowthRate, int totalEpochs, int? seed) Creates a self-paced learning sampler that adapts based on model performance.
SelfPacedSampler<double> SelfPaced(int datasetSize, double initialLambda, double lambdaGrowthRate, int totalEpochs, int? seed) Creates a self-paced learning sampler that adapts based on model performance.
SequentialSampler Sequential(int dataSize) Creates a sequential sampler that iterates through data in order.
StratifiedSampler Stratified(IReadOnlyList<int> labels, int numClasses, int? seed) Creates a stratified sampler that maintains class proportions in each batch.
SubsetSampler Subset(IEnumerable<int> indices, bool shuffle, int? seed) Creates a subset sampler that samples from specific indices.
WeightedSampler<T> Weighted(IEnumerable<T> weights, int numSamples, bool replacement, int? seed) Creates a weighted sampler that samples based on per-sample weights.
WeightedSampler<double> Balanced(IReadOnlyList<int> labels, int numClasses, int? seed) Creates a balanced sampler that oversamples minority classes.