ContinualEvaluationResult<T>

Models & Types · AiDotNet.ContinualLearning.Results

Comprehensive evaluation result across all learned tasks.

For Beginners

This class provides a complete picture of how well the model performs across all tasks it has learned. It includes key metrics for measuring continual learning effectiveness.

How It Works

Key Metrics Explained:

  • Average Accuracy: Mean accuracy across all tasks (higher is better)
  • Backward Transfer: How learning new tasks affects old task performance (positive = improvement, negative = forgetting)
  • Forward Transfer: How old knowledge helps with new tasks (positive = positive transfer)
  • Forgetting: Maximum accuracy drop on any previous task (lower is better)

Reference: Lopez-Paz and Ranzato “Gradient Episodic Memory for Continual Learning” (2017)

Constructors

Constructor Summary
ContinualEvaluationResult(Vector<T> taskAccuracies, Vector<T> taskLosses, T averageAccuracy, T averageLoss, T backwardTransfer, T forwardTransfer, TimeSpan evaluationTime) Initializes a new instance of the ContinualEvaluationResult class.

Properties

Property Summary
IReadOnlyList<TaskEvaluationResult<T>> PerTaskResults { get; set; } Gets the per-task evaluation results.
Matrix<T> AccuracyMatrix { get; set; } Gets the accuracy matrix R where R[i,j] is accuracy on task j after learning task i.
T AverageAccuracy { get; } Gets the average accuracy across all tasks.
T AverageLoss { get; } Gets the average loss across all tasks.
T BackwardTransfer { get; } Gets the backward transfer metric.
T ForwardTransfer { get; } Gets the forward transfer metric.
T Intransigence { get; set; } Gets the intransigence metric (inability to learn new tasks).
T LearningEfficiency { get; set; } Gets the learning curve efficiency (area under the learning curve).
T MaxForgetting { get; set; } Gets the maximum forgetting observed on any task.
T StabilityPlasticityRatio { get; set; } Gets the stability-plasticity ratio.
TimeSpan EvaluationTime { get; } Gets the time taken for evaluation.
Vector<T> TaskAccuracies { get; } Gets the accuracy on each task.
Vector<T> TaskLosses { get; } Gets the loss on each task.
int TaskCount { get; } Gets the total number of tasks evaluated.

Methods

Method Summary
string GenerateReport() Generates a detailed report of the evaluation results.
string ToString() Returns a string representation of the evaluation result.