TaskEvaluationResult<T>

Models & Types · AiDotNet.ContinualLearning.Results

Result from evaluating model performance on a single task.

For Beginners

This class captures how well the model performs on a specific task at a point in time. It’s used to track whether the model remembers what it learned.

Constructors

Constructor Summary
TaskEvaluationResult(int taskId, T accuracy, T loss) Initializes a new instance of the TaskEvaluationResult class.

Properties

Property Summary
IReadOnlyDictionary<ValueTuple<int, int>, int> ConfusionMatrix { get; set; } Gets the confusion matrix, if available.
IReadOnlyDictionary<int, T> PerClassAccuracy { get; set; } Gets per-class accuracy breakdown, if available.
IReadOnlyDictionary<string, T> AdditionalMetrics { get; set; } Gets additional metrics like F1-score, precision, recall.
T Accuracy { get; } Gets the accuracy on this task.
T Loss { get; } Gets the loss on this task.
TimeSpan? EvaluationTime { get; set; } Gets the evaluation time.
Vector<T> ConfidenceScores { get; set; } Gets the confidence scores for predictions, if available.
int TaskId { get; } Gets the task identifier (0-indexed).
int? CorrectCount { get; set; } Gets the number of correct predictions.
int? SampleCount { get; set; } Gets the number of samples evaluated.

Methods

Method Summary
string ToString() Returns a string representation of the evaluation result.