pub fn calibrate(
report: &EvaluationReport,
question: &str,
language: Option<&str>,
risk: Option<&str>,
target_accuracy: f64,
) -> Result<CalibrationResult, EvaluationError>Expand description
Select a question-specific threshold only when the 95% lower confidence bound meets the requested accepted-accuracy target on enough examples.
This returns advice; it never edits runtime configuration.
§Errors
Returns EvaluationError for an invalid report, target, or group.