Struct InferenceGraph 

pub struct InferenceGraph {
    pub nodes: Vec<NodeLatency>,
    pub edges: Vec<(NodeId, NodeId)>,
}

A lightweight Directed Acyclic Graph (DAG) of inference nodes.

Nodes are added in order and receive consecutive NodeIds starting from zero. Edges encode data-flow dependencies: edge (from, to) means “node from must execute before node to”.

Fields

nodes: Vec<NodeLatency>

Per-node latency annotations; index == NodeId.

edges: Vec<(NodeId, NodeId)>

Directed edges (from, to) — i.e. from → to.

Implementations

impl InferenceGraph

pub fn new() -> Self

Create an empty graph.

pub fn add_node(&mut self, latency: NodeLatency) -> NodeId

Add a node with the given latency annotation and return its NodeId.

pub fn add_edge( &mut self, from: NodeId, to: NodeId, ) -> Result<(), CriticalPathError>

Add a directed edge from → to.

Returns CriticalPathError::InvalidNode if either node index is out of range.

pub fn num_nodes(&self) -> usize

Number of nodes in the graph.

pub fn num_edges(&self) -> usize

Number of edges in the graph.

Trait Implementations

impl Clone for InferenceGraph

fn clone(&self) -> InferenceGraph

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for InferenceGraph

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

Formats the value using the given formatter.

impl Default for InferenceGraph

fn default() -> InferenceGraph

Returns the “default value” for a type.