LogScoreMetric<T>
Models & Types · AiDotNet.Evaluation.Metrics.Probabilistic
Logarithmic scoring metric for probabilistic predictions (Negative Log Likelihood).
Implements: IRegressionMetric<T>
For Beginners
This metric measures how well your predicted probability distributions match reality. A lower score means better predictions. It’s particularly useful for models that output full distributions rather than just point predictions.
How It Works
Computes the mean negative log likelihood across all predictions. This is the most common metric for evaluating probabilistic forecasts.
Constructors
| Constructor | Summary |
|---|---|
LogScoreMetric() |
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 log score for probabilistic predictions. |