Struct NeuralDiffusionSchemaConfig 

pub struct NeuralDiffusionSchemaConfig {
    pub hidden_dims: Vec<usize>,
    pub timestep_embed_dim: usize,
    pub learning_rate: f64,
    pub training_epochs: usize,
    pub batch_size: usize,
    pub hybrid_weight: f64,
    pub hybrid_strategy: String,
    pub neural_columns: Vec<String>,
}

Neural diffusion backend configuration.

Controls the candle-based neural score network that learns joint distributions from training data for the neural and hybrid diffusion backends.

Fields

hidden_dims: Vec<usize>

Hidden layer dimensions for the score network MLP.

timestep_embed_dim: usize

Dimensionality of the timestep embedding.

learning_rate: f64

Learning rate for training.

training_epochs: usize

Number of training epochs.

batch_size: usize

Training batch size.

hybrid_weight: f64

Blend weight for hybrid mode (0.0 = all statistical, 1.0 = all neural).

hybrid_strategy: String

Hybrid blending strategy: “weighted_average”, “column_select”, “threshold”.

neural_columns: Vec<String>

Columns to apply neural generation to (empty = all numeric columns).

Trait Implementations

impl Clone for NeuralDiffusionSchemaConfig

fn clone(&self) -> NeuralDiffusionSchemaConfig

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for NeuralDiffusionSchemaConfig

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

Formats the value using the given formatter.

impl Default for NeuralDiffusionSchemaConfig

fn default() -> Self

Returns the “default value” for a type.

impl<'de> Deserialize<'de> for NeuralDiffusionSchemaConfig

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for NeuralDiffusionSchemaConfig

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.