CalibratedProbabilityFitDetector<T, TInput, TOutput>

Models & Types · AiDotNet.FitDetectors

A fit detector that analyzes the calibration of probability predictions to assess model fit.

Inheritance: FitDetectorBase<T, TInput, TOutput> → CalibratedProbabilityFitDetector<T, TInput, TOutput>

For Beginners

Probability calibration refers to how well a model’s predicted probabilities match the actual frequencies of events. A well-calibrated model should predict probabilities that match the true likelihood of events.

How It Works

For example, if a model predicts a 70% probability for 100 different samples, approximately 70 of those samples should actually belong to the positive class. This detector analyzes how well your model’s probability predictions match the actual outcomes.

Constructors

Constructor Summary
CalibratedProbabilityFitDetector(CalibratedProbabilityFitDetectorOptions options) Initializes a new instance of the CalibratedProbabilityFitDetector class.

Methods

Method Summary
FitDetectorResult<T> DetectFit(ModelEvaluationData<T, TInput, TOutput> evaluationData) Detects the fit type of a model based on probability calibration analysis.