IRegressionMetric<T>

Interfaces · AiDotNet.Evaluation.Metrics

Interface for regression metrics.

Implements: IMetric<T>

Known implementations: AdjustedR2Metric<T>, CRPSMetric<T>, ExplainedVarianceMetric<T>, HuberLossMetric<T>, LogCoshLossMetric<T>, LogScoreMetric<T>, MAEMetric<T>, MAPEMetric<T>, MSEMetric<T>, MaxErrorMetric<T>, MeanAbsolutePercentageErrorMetric<T>, MeanBiasErrorMetric<T>, MeanDirectionalAccuracyMetric<T>, MeanSquaredLogErrorMetric<T>, MedianAbsoluteErrorMetric<T>, NormalizedMSEMetric<T>, PearsonCorrelationMetric<T>, PoissonDevianceMetric<T>, QuantileLossMetric<T>, R2ScoreMetric<T>, RMSEMetric<T>, RMSLEMetric<T>, RelativeAbsoluteErrorMetric<T>, RelativeSquaredErrorMetric<T>, SpearmanCorrelationMetric<T>, SymmetricMAPEMetric<T>, TweedieLossMetric<T>, WeightedMAPEMetric<T>

Methods

Method Summary
MetricWithCI<T> ComputeWithCI(ReadOnlySpan<T> predictions, ReadOnlySpan<T> actuals, ConfidenceIntervalMethod ciMethod, double confidenceLevel, int bootstrapSamples, int? randomSeed) Computes the metric with confidence interval.
T Compute(ReadOnlySpan<T> predictions, ReadOnlySpan<T> actuals) Computes the metric from predicted and actual values.