Struct OwnedSeparableCellMeasure 

pub struct OwnedSeparableCellMeasure { /* private fields */ }

Lifetime-free structural geometry and likelihood measure.

The realized (n_rows, n_outputs) shape is part of the value. This makes a stored measure self-validating and prevents reusing a sparse activity set or weight vector against a different response grid.

Implementations

impl OwnedSeparableCellMeasure

pub fn new( n_rows: usize, n_outputs: usize, structural: OwnedStructuralCells, likelihood_weights: OwnedLikelihoodWeights, ) -> Result<Self, IndexedResponseError>

Construct and validate an owned response measure.

pub fn uniform(n_rows: usize, n_outputs: usize) -> Self

All cells present with unit numerical weight.

pub fn n_rows(&self) -> usize

pub fn n_outputs(&self) -> usize

pub fn as_borrowed(&self) -> SeparableCellMeasure<'_>

Borrow this owned value through the zero-copy evaluation API.

pub fn is_active(&self, row: usize, output: usize) -> bool

pub fn active_weight(&self, row: usize, output: usize) -> Option<f64>

pub fn try_for_each_active<E>( &self, visitor: impl FnMut(usize, usize, f64) -> Result<(), E>, ) -> Result<(), E>

Visit every structurally active cell in deterministic row-major order. Sparse inclusion geometry runs in O(active cells); zero numerical weights remain visible to the visitor because they do not erase model structure.

Trait Implementations

impl Clone for OwnedSeparableCellMeasure

fn clone(&self) -> OwnedSeparableCellMeasure

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for OwnedSeparableCellMeasure

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

Formats the value using the given formatter.

impl PartialEq for OwnedSeparableCellMeasure

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

Equality operator ==.

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

Inequality operator !=.

impl StructuralPartialEq for OwnedSeparableCellMeasure