QuantileLossMetric<T>

Models & Types · AiDotNet.Evaluation.Metrics.Regression

Computes Quantile Loss (Pinball Loss) for quantile regression.

Implements: IRegressionMetric<T>

For Beginners

Quantile loss is used when you want to predict a specific percentile:

  • τ = 0.5: Median (equivalent to MAE)
  • τ = 0.9: 90th percentile (over-predictions penalized less)
  • τ = 0.1: 10th percentile (under-predictions penalized less)

How It Works

Quantile Loss = (1/N) × Σ max(τ(y - ŷ), (τ - 1)(y - ŷ))

Use cases:

  • Risk assessment (predict 95th percentile of losses)
  • Inventory planning (predict 99th percentile of demand)
  • Uncertainty quantification (predict multiple quantiles)

Constructors

Constructor Summary
QuantileLossMetric(double quantile)

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)