Struct ReasoningEvalSummary 

pub struct ReasoningEvalSummary {
    pub num_benchmarks: usize,
    pub accuracy: HashMap<String, f64>,
    pub improvement: HashMap<String, f64>,
    pub self_consistency: f64,
    pub calibration_ece: f64,
    pub thinktool_scores: HashMap<String, f64>,
    /* private fields */
}

Summary of reasoning evaluation

Fields

num_benchmarks: usize

accuracy: HashMap<String, f64>

Accuracy per benchmark

improvement: HashMap<String, f64>

Improvement over baseline per benchmark

self_consistency: f64

Self-consistency rate

calibration_ece: f64

Expected Calibration Error

thinktool_scores: HashMap<String, f64>

ThinkTool effectiveness scores

Implementations

impl ReasoningEvalSummary

pub fn new(benchmarks: &[String]) -> Self

pub fn check_targets(&self, targets: &ReasoningTargets) -> TargetResult

Check if reasoning targets are met

Trait Implementations

impl Clone for ReasoningEvalSummary

fn clone(&self) -> ReasoningEvalSummary

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for ReasoningEvalSummary

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.