SineWaveMetaDataset<T, TInput, TOutput>

Models & Types · AiDotNet.MetaLearning.Data

A synthetic meta-dataset for regression where each task is a sinusoidal function with random amplitude and phase. This is the standard benchmark from the MAML paper (Finn et al., ICML 2017).

Inheritance: MetaDatasetBase<T, TInput, TOutput> → SineWaveMetaDataset<T, TInput, TOutput>

For Beginners

This dataset generates sine wave tasks like y = A * sin(x + phase). Each “class” is defined by a unique (amplitude, phase) pair. The meta-learner must learn to quickly fit a sine wave from just a few (x, y) points.

Constructors

Constructor Summary
SineWaveMetaDataset(int numClasses, int examplesPerClass, double amplitudeMin, double amplitudeMax, double phaseMin, double phaseMax, double xMin, double xMax, int? seed) Creates a sine wave meta-dataset.

Properties

Property Summary
IReadOnlyDictionary<int, int> ClassExampleCounts { get; }
int TotalClasses { get; }
int TotalExamples { get; }
string Name { get; }