Struct GNNInference 

pub struct GNNInference { /* private fields */ }

GNN Inference kernel.

Performs message passing neural network inference on graph data. Supports various aggregation strategies and can be used for node classification, link prediction, and graph-level tasks.

Implementations

impl GNNInference

pub fn new() -> Self

Create a new GNN Inference kernel.

pub fn compute( graph: &CsrGraph, node_features: &[Vec<f64>], weights: &GNNWeights, config: &GNNConfig, ) -> GNNResult

Run GNN inference on a graph.

Trait Implementations

impl Clone for GNNInference

fn clone(&self) -> GNNInference

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for GNNInference

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

Formats the value using the given formatter.

impl Default for GNNInference

fn default() -> Self

Returns the “default value” for a type.

impl GpuKernel for GNNInference

fn metadata(&self) -> &KernelMetadata

Returns the kernel metadata.

fn validate(&self) -> Result<(), KernelError>

Validate kernel configuration.

fn id(&self) -> &str

Returns the kernel ID.

fn requires_gpu_native(&self) -> bool

Returns true if this kernel requires GPU-native execution.

fn health_check(&self) -> HealthStatus

Perform a health check on this kernel.

fn shutdown(&self) -> Result<(), KernelError>

Graceful shutdown of the kernel.

fn refresh_config(&mut self, _config: &KernelConfig) -> Result<(), KernelError>

Refresh kernel configuration at runtime.