Struct PrCurve 

pub struct PrCurve {
    pub points: Vec<PrPoint>,
    pub average_precision: f64,
    pub baseline: f64,
}

A precision–recall curve: operating points, average precision, and the positive-class prevalence (the curve’s chance baseline).

Fields

points: Vec<PrPoint>

Operating points, ascending in recall.

average_precision: f64

Average precision — the step-function area under the PR curve, using the sum_n (R_n - R_{n-1}) * P_n definition (matching scikit-learn’s average_precision_score), not a trapezoidal interpolation.

baseline: f64

Positive-class prevalence, P / (P + N). A no-skill classifier’s PR curve is a horizontal line at this height — the correct PR baseline (a diagonal, which is the ROC baseline, would be wrong here).

Trait Implementations

impl Clone for PrCurve

fn clone(&self) -> PrCurve

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for PrCurve

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

Formats the value using the given formatter.

impl PartialEq for PrCurve

fn eq(&self, other: &PrCurve) -> bool

Equality operator ==.

fn ne(&self, other: &Rhs) -> bool

Inequality operator !=.

impl StructuralPartialEq for PrCurve