CRPSMetric<T>

Models & Types · AiDotNet.Evaluation.Metrics.Probabilistic

Continuous Ranked Probability Score metric for probabilistic predictions.

Implements: IRegressionMetric<T>

For Beginners

CRPS tells you how good your probability forecasts are. Unlike log score, CRPS:

  • Has the same units as your predicted variable
  • For point predictions, equals mean absolute error
  • Rewards both accuracy and appropriate uncertainty

Lower CRPS is better.

How It Works

CRPS is a proper scoring rule that measures the quality of probabilistic predictions. It generalizes mean absolute error to full probability distributions.

Constructors

Constructor Summary
CRPSMetric()

Properties

Property Summary
MetricDirection Direction { get; }
T MaxValue { get; }
T MinValue { get; }
bool RequiresProbabilities { get; }
bool SupportsMultiClass { get; }
string Category { get; }
string Description { get; }
string Name { get; }

Methods

Method Summary
MetricWithCI<T> ComputeWithCI(ReadOnlySpan<T> predictions, ReadOnlySpan<T> actuals, ConfidenceIntervalMethod ciMethod, double confidenceLevel, int bootstrapSamples, int? randomSeed)
T Compute(ReadOnlySpan<T> predictions, ReadOnlySpan<T> actuals)
T ComputeFromDistributions(IParametricDistribution<T>[] distributions, Vector<T> observations) Computes the mean CRPS for probabilistic predictions.