Struct MemoryGraph 

pub struct MemoryGraph { /* private fields */ }

A memory link graph for computing PageRank-style importance.

Nodes are memory entry IDs (mapped to u64 indices internally). Edges represent co-access relationships.

Implementations

impl MemoryGraph

pub fn new() -> Self

Create an empty graph.

pub fn add_edge(&mut self, from: u64, to: u64)

Add a directed edge from from to to.

If the edge already exists, this is a no-op.

pub fn link(&mut self, a: u64, b: u64)

Add a bidirectional link between two nodes (co-access).

pub fn node_count(&self) -> usize

Get the number of nodes in the graph.

pub fn neighbors(&self, node: u64) -> &[u64]

Get outgoing neighbors of a node.

pub fn pagerank( &self, damping: f64, iterations: usize, initial_scores: Option<&HashMap<u64, f64>>, ) -> HashMap<u64, f64>

Compute PageRank scores for all nodes.

Uses the standard iterative algorithm with damping factor.

Arguments

• damping — Damping factor (typically 0.85).
• iterations — Number of iterations (typically 20-50).
• initial_scores — Optional initial scores (e.g., base importance). If provided, the initial PageRank is seeded with these values.

Returns

A map of node ID -> PageRank score.

pub fn from_co_access(sessions: &[Vec<u64>]) -> Self

Compute co-access graph from session access patterns.

Given a list of sessions where each session contains a list of memory IDs that were accessed together, build a graph where co-accessed memories are linked.

Trait Implementations

impl Clone for MemoryGraph

fn clone(&self) -> MemoryGraph

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for MemoryGraph

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

Formats the value using the given formatter.

impl Default for MemoryGraph

fn default() -> MemoryGraph

Returns the “default value” for a type.