Package org.egothor.methodatlas.eval

package org.egothor.methodatlas.eval

AI-classification accuracy evaluation: scores AI predictions against a human-reviewed labelled set (the ClassificationOverride YAML format) and reports precision, recall, F1 and support per taxonomy tag and for the securityRelevant decision.

AccuracyMetrics is the pure metric math, AccuracyEvaluator joins labels with predictions, and AccuracyReportRenderer renders Markdown or JSON. Nothing in this package calls an AI provider.

Related Packages

Package	Description
org.egothor.methodatlas	Provides the core command-line utility for analyzing Java test sources and producing structured metadata about JUnit test methods.

Class	Description
AccuracyEvaluator	Joins AI predictions with human-reviewed labels and scores them with AccuracyMetrics.
AccuracyEvaluator.Evaluation	Metrics together with the coverage bookkeeping of the join.
AccuracyMetrics	Pure, deterministic computation of AI-classification quality metrics from (prediction, truth) pairs.
AccuracyMetrics.Counts	Confusion counts for one binary decision.
AccuracyMetrics.Result	The complete metric set.
AccuracyMetrics.Sample	One method for which a prediction is compared with a human label.
AccuracyMetrics.TagResult	Metrics of a single taxonomy tag.
AccuracyReportRenderer	Renders an AccuracyEvaluator.Evaluation as Markdown or JSON.