CalibrationResult<T>

Models & Types · AiDotNet.Evaluation.Engines

Results from calibration analysis.

Constructors

Constructor Summary
CalibrationResult()

Properties

Property Summary
List<CalibrationBin<T>> Bins { get; set; } Calibration bins with predicted vs observed frequencies.
T BrierScore { get; set; } Brier Score: mean squared error of probability predictions.
T ExpectedCalibrationError { get; set; } Expected Calibration Error: weighted average of |predicted - observed| across bins.
T MaximumCalibrationError { get; set; } Maximum Calibration Error: worst bin’s |predicted - observed|.
bool IsWellCalibrated { get; set; } Whether the classifier is considered well-calibrated (ECE < 0.05).
int NumSamples { get; set; } Total number of samples analyzed.