IDifficultyEstimator<T, TInput, TOutput>
Interfaces · AiDotNet.CurriculumLearning.Interfaces
Interface for estimating the difficulty of training samples.
Known implementations: DifficultyEstimatorBase<T, TInput, TOutput>, IConfidentDifficultyEstimator<T, TInput, TOutput>, IEnsembleDifficultyEstimator<T, TInput, TOutput>
For Beginners
A difficulty estimator measures how “hard” each training sample is for the model to learn. This is crucial for curriculum learning, as we want to present easy samples first and gradually introduce harder ones.
How It Works
Common Difficulty Measures:
Research Background:
Properties
| Property | Summary |
|---|---|
bool RequiresModel { get; } |
Gets whether this estimator requires the model to estimate difficulty. |
string Name { get; } |
Gets the name of the difficulty estimator. |
Methods
| Method | Summary |
|---|---|
T EstimateDifficulty(TInput input, TOutput expectedOutput, IFullModel<T, TInput, TOutput> model) |
Estimates the difficulty of a single sample. |
Vector<T> EstimateDifficulties(IDataset<T, TInput, TOutput> dataset, IFullModel<T, TInput, TOutput> model) |
Estimates difficulty scores for all samples in a dataset. |
int[] GetSortedIndices(Vector<T> difficulties) |
Gets the indices of samples sorted by difficulty (easy to hard). |
void Reset() |
Resets the estimator to its initial state. |
void Update(int epoch, IFullModel<T, TInput, TOutput> model) |
Updates the difficulty estimator based on training progress. |