Struct LargeScaleVariationalConfig 

pub struct LargeScaleVariationalConfig {
    pub max_epochs: usize,
    pub batch_size: usize,
    pub learning_rate: Float,
    pub learning_rate_decay: LearningRateDecay,
    pub tolerance: Float,
    pub n_mc_samples: usize,
    pub use_natural_gradients: bool,
    pub use_control_variates: bool,
    pub memory_limit_gb: Option<Float>,
    pub verbose: bool,
    pub random_seed: Option<u64>,
    pub prior_config: PriorConfiguration,
}

Configuration for large-scale variational inference

Fields

max_epochs: usize

Maximum number of epochs

batch_size: usize

Mini-batch size for stochastic updates

learning_rate: Float

Learning rate for variational parameter updates

learning_rate_decay: LearningRateDecay

Learning rate decay schedule

tolerance: Float

Convergence tolerance

n_mc_samples: usize

Number of Monte Carlo samples for expectations

use_natural_gradients: bool

Whether to use natural gradients

use_control_variates: bool

Whether to use control variates for variance reduction

memory_limit_gb: Option<Float>

Memory limit in GB for adaptive batch sizing

verbose: bool

Whether to enable verbose output

random_seed: Option<u64>

Random seed for reproducibility

prior_config: PriorConfiguration

Prior parameters

Trait Implementations

impl Clone for LargeScaleVariationalConfig

fn clone(&self) -> LargeScaleVariationalConfig

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for LargeScaleVariationalConfig

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

Formats the value using the given formatter.

impl Default for LargeScaleVariationalConfig

fn default() -> Self

Returns the “default value” for a type.