Struct GatConfig 

pub struct GatConfig {
    pub input_dim: usize,
    pub head_output_dim: usize,
    pub num_heads: usize,
    pub dropout: f64,
    pub alpha: f64,
    pub concat_heads: bool,
    pub normalize_output: bool,
    pub seed: u64,
}

Configuration for a Graph Attention Network

Fields

input_dim: usize

Dimensionality of input node features

head_output_dim: usize

Dimensionality of output per attention head

num_heads: usize

Number of attention heads

dropout: f64

Dropout rate (applied to attention coefficients)

alpha: f64

LeakyReLU negative slope for attention scoring

concat_heads: bool

If true, concatenate head outputs; else average them

normalize_output: bool

L2-normalize output embeddings

seed: u64

Random seed for parameter initialization

Implementations

impl GatConfig

pub fn output_dim(&self) -> usize

Compute the final output dimensionality

Trait Implementations

impl Clone for GatConfig

fn clone(&self) -> GatConfig

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for GatConfig

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

Formats the value using the given formatter.

impl Default for GatConfig

fn default() -> Self

Returns the “default value” for a type.

impl<'de> Deserialize<'de> for GatConfig

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

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for GatConfig

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

where __S: Serializer,

Serialize this value into the given Serde serializer.