Struct GammaPoissonObservation 

pub struct GammaPoissonObservation {
    pub rate_variable: String,
    pub observations: Vec<u64>,
}

GammaPoissonObservation captures a Poisson likelihood y ~ Poisson(λ) where the rate λ is a GammaNP variable. It contributes (Σ y_i, N) to the posterior natural parameters, i.e. adds Σ y_i to (α − 1) and N to −(−β) = β.

A factor holds a reference to its Gamma-distributed rate variable and a (possibly empty) batch of observations. Posterior inference combining prior + factor is exact in one VMP sweep because Poisson is conjugate to Gamma.

Fields

rate_variable: String

Name of the GammaNP variable in the VMP graph.

observations: Vec<u64>

Observed Poisson counts.

Implementations

impl GammaPoissonObservation

pub fn new(rate_variable: impl Into<String>, observations: Vec<u64>) -> Self

Build a new Gamma-Poisson observation factor.

pub fn count_sum(&self) -> u64

Sum of observed counts Σ y_i. Used as the shape-parameter increment.

pub fn num_observations(&self) -> usize

Number of observations N. Used as the rate-parameter increment.

Trait Implementations

impl Clone for GammaPoissonObservation

fn clone(&self) -> GammaPoissonObservation

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for GammaPoissonObservation

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

Formats the value using the given formatter.