Struct MeanFieldGaussian 

pub struct MeanFieldGaussian {
    pub means: Array1<f64>,
    pub log_stds: Array1<f64>,
    pub dim: usize,
}

Mean-field Gaussian variational family

Approximates the posterior with a factorized Gaussian: q(z) = prod_i N(z_i; mu_i, sigma_i^2)

This is the simplest variational family but cannot capture correlations.

Fields

means: Array1<f64>

Variational means

log_stds: Array1<f64>

Variational log standard deviations (unconstrained parameterization)

dim: usize

Dimensionality

Implementations

impl MeanFieldGaussian

pub fn new(dim: usize) -> Result<Self>

Create a new mean-field Gaussian with given dimension

pub fn from_params(means: Array1<f64>, log_stds: Array1<f64>) -> Result<Self>

Create from specific parameters

pub fn stds(&self) -> Array1<f64>

Get the standard deviations

pub fn variances(&self) -> Array1<f64>

Get the variances

pub fn sample(&self, epsilon: &Array1<f64>) -> Result<Array1<f64>>

Sample from the variational distribution using reparameterization trick

z = mu + sigma * epsilon, where epsilon ~ N(0, I)

pub fn entropy(&self) -> f64

Compute the entropy of the mean-field Gaussian H[q] = sum_i 0.5 * (1 + log(2pi) + 2log_std_i)

pub fn log_prob(&self, z: &Array1<f64>) -> Result<f64>

Compute log q(z) for a given z

pub fn n_params(&self) -> usize

Get total number of variational parameters

pub fn get_params(&self) -> Array1<f64>

Get all variational parameters as a flat vector

pub fn set_params(&mut self, params: &Array1<f64>) -> Result<()>

Set variational parameters from a flat vector

Trait Implementations

impl Clone for MeanFieldGaussian

fn clone(&self) -> MeanFieldGaussian

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for MeanFieldGaussian

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

Formats the value using the given formatter.