Struct MultivariateNormalGibbs 

pub struct MultivariateNormalGibbs {
    pub mean: Array1<f64>,
    pub precision: Array2<f64>,
}

Multivariate normal Gibbs sampler

For sampling from a multivariate normal distribution where each variable given all others follows a normal distribution.

Fields

mean: Array1<f64>

Mean vector

precision: Array2<f64>

Precision matrix (inverse covariance)

Implementations

impl MultivariateNormalGibbs

pub fn new(mean: Array1<f64>, covariance: Array2<f64>) -> Result<Self>

Create a new multivariate normal Gibbs sampler

pub fn from_precision(mean: Array1<f64>, precision: Array2<f64>) -> Result<Self>

Create from precision matrix directly

Trait Implementations

impl Clone for MultivariateNormalGibbs

fn clone(&self) -> MultivariateNormalGibbs

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl ConditionalDistribution for MultivariateNormalGibbs

fn sample_conditional<R: Rng + ?Sized>( &self, current_state: &Array1<f64>, variable_index: usize, rng: &mut R, ) -> Result<f64>

Sample from the conditional distribution P(X_i | X_{-i})

fn dim(&self) -> usize

Get the dimensionality of the distribution

fn log_density(&self, x: &Array1<f64>) -> Option<f64>

Optionally compute log density for monitoring

impl Debug for MultivariateNormalGibbs

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

Formats the value using the given formatter.