CalibrationEngine class

Calibration analysis utilities.

Compares a learner's self-reported confidence with their actual performance to detect overconfidence and underconfidence patterns.

Constructors

CalibrationEngine()

Properties

hashCode → int
The hash code for this object.
no setterinherited
runtimeType → Type
A representation of the runtime type of the object.
no setterinherited

Methods

noSuchMethod(Invocation invocation) → dynamic
Invoked when a nonexistent method or property is accessed.
inherited
toString() → String
A string representation of this object.
inherited

Operators

operator ==(Object other) → bool
The equality operator.
inherited

Static Methods

calibrationCurve(List<ConfidenceEvent> events) → Map<int, double>
Calibration curve: for each confidence level (1-5), what is the average actual performance? Returns a map: {1: 0.32, 2: 0.45, ...}.
calibrationError(List<ConfidenceEvent> events) → double
Mean absolute error between predicted (normalized to 0-1) and actual.
forTopic(List<ConfidenceEvent> events, String topicId) → List<ConfidenceEvent>
Filter events for a specific topic.
isOverconfident(List<ConfidenceEvent> events, {double threshold = 0.25}) → bool
Whether the learner is overconfident on a given topic.