Struct NeuralPredDecl 

pub struct NeuralPredDecl {
    pub network: String,
    pub inputs: Vec<String>,
    pub output: String,
    pub labels: Option<Vec<NeuralLabel>>,
    pub predicate: Atom,
}

Neural predicate declaration

Neural predicates connect neural networks to probabilistic logic. Syntax: nn(network, [inputs], output, [labels]) :: pred(args).

The neural network produces probability distributions over labels, which become probabilistic facts in the logic program.

Examples

nn(mnist_net, [X], Y, [0,1,2,3,4,5,6,7,8,9]) :: digit(X, Y).
nn(encoder, [Text], Embedding) :: encode(Text, Embedding).

Fields

network: String

Name of the registered neural network

inputs: Vec<String>

Input variable names (bind to tensor sources)

output: String

Output variable name

labels: Option<Vec<NeuralLabel>>

Optional classification labels (for classification networks) If None, the network produces embeddings

predicate: Atom

The predicate this neural network defines

Trait Implementations

impl Clone for NeuralPredDecl

fn clone(&self) -> NeuralPredDecl

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for NeuralPredDecl

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

Formats the value using the given formatter.

impl PartialEq for NeuralPredDecl

fn eq(&self, other: &NeuralPredDecl) -> bool

Tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl StructuralPartialEq for NeuralPredDecl