TransferBasedDifficultyEstimator<T, TInput, TOutput>

Models & Types · AiDotNet.CurriculumLearning.DifficultyEstimators

Difficulty estimator based on transfer learning from a simpler “teacher” model.

Inheritance: DifficultyEstimatorBase<T, TInput, TOutput> → TransferBasedDifficultyEstimator<T, TInput, TOutput>

For Beginners

This estimator uses the performance gap between a simple “teacher” model and the main “student” model to estimate difficulty. Samples that are easy for the simple model but hard for the main model are considered easier, while samples that are hard for both are considered more difficult.

How It Works

How It Works:

Difficulty Calculation:

References:

Constructors

Constructor Summary
TransferBasedDifficultyEstimator(IFullModel<T, TInput, TOutput> teacherModel, TransferDifficultyMode mode, bool normalize) Initializes a new instance of the TransferBasedDifficultyEstimator class.

Properties

Property Summary
IFullModel<T, TInput, TOutput> TeacherModel { get; } Gets the teacher model used for comparison.
bool RequiresModel { get; } Gets whether this estimator requires the main model.
string Name { get; } Gets the name of this estimator.

Methods

Method Summary
T EstimateDifficulty(TInput input, TOutput expectedOutput, IFullModel<T, TInput, TOutput> model) Estimates the difficulty of a single sample using transfer comparison.
Vector<T> EstimateDifficulties(IDataset<T, TInput, TOutput> dataset, IFullModel<T, TInput, TOutput> model) Estimates difficulty scores for all samples.