FairnessResult<T>

Models & Types · AiDotNet.Evaluation.Engines

Results from fairness analysis.

Constructors

Constructor Summary
FairnessResult()

Properties

Property Summary
Dictionary<int, GroupFairnessMetrics> GroupMetrics { get; set; } Per-group fairness metrics.
bool IsFair { get; set; } Whether the model is considered fair based on configured thresholds.
double AverageOddsDifference { get; set; } Average Odds Difference: mean of TPR and FPR differences.
double DemographicParityDifference { get; set; } Demographic Parity Difference: max - min positive prediction rate across groups.
double DisparateImpactRatio { get; set; } Disparate Impact Ratio: min positive rate / max positive rate.
double EqualOpportunityDifference { get; set; } Equal Opportunity Difference: max - min TPR across groups.
double EqualizedOddsDifference { get; set; } Equalized Odds Difference: max of TPR and FPR differences across groups.
double TheilIndex { get; set; } Theil Index: measures inequality in model benefit distribution.
int NumSamples { get; set; } Total number of samples analyzed.
int[] Groups { get; set; } Array of group identifiers.