Struct HypergraphAttentionLayer 

pub struct HypergraphAttentionLayer { /* private fields */ }

A single Hypergraph Attention layer.

Processes node features through two rounds of dual-direction attention:

1. Nodes → Hyperedges (aggregate node info into hyperedge representations)
2. Hyperedges → Nodes (aggregate hyperedge info back to node representations)

Implementations

impl HypergraphAttentionLayer

pub fn new(in_dim: usize, config: HypergraphAttentionConfig) -> Self

Create a new HypergraphAttentionLayer.

Arguments

• in_dim: input node feature dimension
• config: layer configuration

pub fn forward( &self, node_feats: &Array2<f64>, incidence_matrix: &Array2<f64>, ) -> Result<Array2<f64>>

Forward pass.

Arguments

• node_feats: node features, shape [N × in_dim]
• incidence_matrix: H ∈ {0,1}^{N×M}, shape [N × M]

Returns

Updated node features, shape [N × in_dim].

Trait Implementations

impl Clone for HypergraphAttentionLayer

fn clone(&self) -> HypergraphAttentionLayer

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for HypergraphAttentionLayer

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

Formats the value using the given formatter.