Struct TrainingSupport 

pub struct TrainingSupport {
    pub axis_min: Array1<f64>,
    pub axis_max: Array1<f64>,
}

Per-axis training support range used by boundary and OOD corrections. For each predictor axis we record the empirical [min, max] from training. Boundary correction inflates variance for x_i within a small fraction of the range from either edge; OOD inflation inflates variance for x_i outside [min, max] proportional to (excess / range).

Fields

axis_min: Array1<f64>

Axis-wise minimum across the training rows; length = number of input columns the design treats as continuous predictors. The order must match predictor_x rows passed in PredictUncertaintyOptions:: predictor_x_for_corrections (see helper below); a length of zero disables both boundary and OOD corrections.

axis_max: Array1<f64>

Axis-wise maximum, paired with axis_min.

Implementations

impl TrainingSupport

pub fn from_training_rows(rows: ArrayView2<'_, f64>) -> Self

Convenience constructor from raw training rows. Computes per-axis min/max in a single pass.

Trait Implementations

impl Clone for TrainingSupport

fn clone(&self) -> TrainingSupport

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for TrainingSupport

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

Formats the value using the given formatter.