Struct CorrectionFit 

pub struct CorrectionFit { /* private fields */ }

A fitted correction: coefficients, bootstrap replicates and the split.

Implementations

impl CorrectionFit

pub fn ion(&self) -> &Ion

The projectile the correction was fitted for.

pub fn target_z(&self) -> u8

Atomic number of the target element.

pub fn span_ev(&self) -> (f64, f64)

Energy span (lowest, highest) of the training points, eV; the correction applies inside it only (both ends included).

pub fn coefficients(&self) -> &[f64]

Chebyshev coefficients c_0..=c_d of ln g.

pub fn settings(&self) -> FitSettings

The settings the fit was made with (including the seed).

pub fn train_indices(&self) -> &[usize]

Dataset indices of the training points, ascending.

pub fn held_out_indices(&self) -> &[usize]

Dataset indices of the held-out points, ascending.

pub fn weighted(&self) -> bool

True if the fit used weights 1 / (sigma / S)^2, which requires every training point to state an uncertainty; otherwise it is unweighted.

pub fn failed_replicates(&self) -> usize

Bootstrap resamples discarded because their fit was singular.

pub fn n_replicates(&self) -> usize

Number of bootstrap replicates kept.

pub fn held_out_report(&self) -> &HeldOutReport

Residuals of the base and corrected model on the held-out points.

pub fn train_base(&self) -> ResidualStats

Residual statistics of the base model on the training points.

pub fn train_corrected(&self) -> ResidualStats

Residual statistics of the corrected model on the training points.

pub fn in_span(&self, energy_ev: f64) -> bool

True if energy_ev is inside the span the correction applies to.

pub fn factor(&self, energy_ev: f64) -> Option<f64>

The correction factor g(E) of the fit, or None outside the span.

pub fn band(&self, energy_ev: f64, level: f64) -> Option<(f64, f64)>

Pointwise bootstrap band of g(E): the central interval holding level (for example 0.68) of the replicates, as (low, high).

None outside the span, with no replicates, or for level outside (0, 1).

Trait Implementations

impl Clone for CorrectionFit

fn clone(&self) -> CorrectionFit

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for CorrectionFit

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

Formats the value using the given formatter.

impl PartialEq for CorrectionFit

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

Equality operator ==.

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

Inequality operator !=.

impl StructuralPartialEq for CorrectionFit