Struct DirichletMultinomial 

pub struct DirichletMultinomial {
    pub alpha: Array1<f64>,
}

Dirichlet-Multinomial conjugate pair

Prior: Dirichlet(α) Likelihood: Multinomial(n, p) Posterior: Dirichlet(α + counts)

Fields

alpha: Array1<f64>

Concentration parameters of the Dirichlet prior

Implementations

impl DirichletMultinomial

pub fn new(alpha: Array1<f64>) -> Result<Self>

Create a new Dirichlet-Multinomial conjugate prior

pub fn uniform(k: usize) -> Result<Self>

Create uniform prior with given dimension

pub fn update(&self, counts: ArrayView1<'_, f64>) -> Result<Self>

Update the prior with observed count data

Arguments

• counts - Array of observed counts for each category

Returns

Updated DirichletMultinomial with posterior parameters

pub fn posterior_mean(&self) -> Result<Array1<f64>>

Compute the posterior mean

pub fn posterior_mode(&self) -> Result<Option<Array1<f64>>>

Compute the posterior mode (MAP estimate)

pub fn posterior_variance(&self) -> Result<Array1<f64>>

Compute the marginal variance for each component

Trait Implementations

impl Clone for DirichletMultinomial

fn clone(&self) -> DirichletMultinomial

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for DirichletMultinomial

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

Formats the value using the given formatter.