Struct NCNPAG 

pub struct NCNPAG<E: Equation + Send + 'static> { /* private fields */ }

Non-collapsing NPAG (NCNPAG) algorithm.

Individualizes a set of prior support points to a subject’s data in two steps, without collapsing (merging) the points:

1. Bayesian filtering. Evaluate the likelihood P(data | θⱼ) for each prior support point, apply a flat (uniform) prior so the posterior weight is proportional to the likelihood (postⱼ ∝ ∏ᵢ P(dataᵢ | θⱼ)), drop points whose normalized weight falls below 1e-100 × max, and renormalize the survivors.
2. Per-point NPAG refinement. When cycles > 0, seed a full NPAG run from each surviving point and replace it with the resulting daughter point, preserving its filter weight. Points whose refinement fails or yields nothing are kept at their original location.

The result is returned as a standard NonParametricResult, so the (theta, weights) can be consumed exactly like any other fit.

Trait Implementations

impl<E: Equation + Send + 'static> NonParametricRunner<E> for NCNPAG<E>

fn fit(&mut self) -> Result<NonParametricResult<E>>

NCNPAG is a single-pass reweighting: it evaluates the likelihood of the fixed prior support points once, rather than iterating cycles.

fn into_result(&self) -> Result<NonParametricResult<E>>

fn error_models(&self) -> &AssayErrorModels

fn equation(&self) -> &E

Get the equation used in the algorithm

fn data(&self) -> &Data

Get the data used in the algorithm

fn likelihood(&self) -> f64

Get the current likelihood

fn increment_cycle(&mut self) -> usize

Increment the cycle counter and return the new value

fn cycle(&self) -> usize

Get the current cycle number

fn set_theta(&mut self, theta: Theta)

Set the current Theta

fn theta(&self) -> &Theta

Get the current Theta

fn psi(&self) -> &Psi

Get the current Psi

fn set_status(&mut self, status: Status)

Set the current Status of the algorithm

fn status(&self) -> &Status

Get the current Status of the algorithm

fn evaluation(&mut self) -> Result<Status>

Evaluate convergence criteria and update status

fn estimation(&mut self) -> Result<()>

fn condensation(&mut self) -> Result<()>

Performs condensation of Theta and updates Psi

fn optimizations(&mut self) -> Result<()>

Performs optimizations on the current AssayErrorModels and updates Psi accordingly

fn expansion(&mut self) -> Result<()>

Performs expansion of Theta

fn log_cycle_state(&mut self)

Create and log a cycle state with the current algorithm state

fn check_zero_probability_subjects(&self) -> Result<()>

Identify subjects whose total probability given the model is zero or non-finite.

fn log_zero_probability_subject(&self, subject: &Subject)

Log detailed likelihood diagnostics for a single subject whose probability given the model is zero or non-finite.

fn n2ll(&self) -> f64

Get the current negative two log-likelihood

fn initialize(&mut self) -> Result<()>

Initialize the algorithm, setting up initial Theta and Status

fn next_cycle(&mut self) -> Result<Status>

Proceed to the next cycle of the algorithm