Struct SynapticGraph 

pub struct SynapticGraph { /* private fields */ }

Adjacency structure for a spiking neural network with delay and polarity metadata.

Stores connections in Compressed Sparse Row (CSR) format, augmented with per-synapse delay and polarity vectors aligned to the same index space.

Layout

row_ptr:    [0,  3,  7, 10, ...]     — start index per source neuron
targets:    [5, 12,  0,  1,  3, ...]  — target neuron IDs
weights:    [0.9, -0.15, 0.3, ...]    — effective signed weights
delays:     [2, 5, 1, ...]            — axonal delay in ticks
polarities: [Exc, Inh, Exc, ...]      — Dale's law polarity

Implementations

impl SynapticGraph

pub fn topology_digest(&self) -> TopologyDigest

Deterministic, versioned digest of this graph’s logical topology.

See TopologyDigest for the schema, domain separator, and float bit-pattern rules. Equivalent graphs built through different insertion orders produce the same value.

use synaptic_wiring::topology::SynapticGraph;
use synaptic_wiring::types::{Polarity, SynapseDescriptor};

let a = [
    SynapseDescriptor {
        source: 0,
        target: 1,
        weight: 0.5,
        delay: 2,
        polarity: Polarity::Excitatory,
    },
    SynapseDescriptor {
        source: 1,
        target: 0,
        weight: 0.25,
        delay: 1,
        polarity: Polarity::Inhibitory,
    },
];
let mut b = a;
b.swap(0, 1);

let ga = SynapticGraph::from_descriptors(2, &a).unwrap();
let gb = SynapticGraph::from_descriptors(2, &b).unwrap();
assert_eq!(ga.topology_digest(), gb.topology_digest());

impl SynapticGraph

pub fn new(n: usize) -> Self

Create an empty graph with n neurons and no connections.

pub fn from_descriptors( neuron_count: usize, descriptors: &[SynapseDescriptor], ) -> Result<Self>

Build from a list of synapse descriptors.

Descriptors need not be sorted; they will be grouped by source neuron. Each descriptor weight must be a finite non-negative magnitude (IEEE signed zero is accepted). Negative, NaN, and infinite magnitudes are rejected — they are not abs()-normalized.

pub fn neuron_count(&self) -> usize

Number of neurons in the graph.

pub fn synapse_count(&self) -> usize

Total number of synapses (non-zero entries).

pub fn sparsity(&self) -> f32

Sparsity: fraction of the N×N matrix that is zero.

pub fn outgoing( &self, source: usize, ) -> impl Iterator<Item = (NeuronId, f32, DelayTicks, Polarity)> + '_

Iterate over all outgoing synapses from source. Returns (target, weight, delay, polarity) tuples.

pub fn out_degree(&self, neuron: usize) -> usize

Get the out-degree of a neuron.

pub fn max_delay(&self) -> DelayTicks

Maximum delay present in the graph.

pub fn mean_degree(&self) -> f32

Mean out-degree across all neurons.

pub fn to_gpu_arrays(&self) -> (Vec<u32>, Vec<u32>, Vec<f32>, Vec<u16>)

Raw access to CSR arrays for GPU upload.

pub fn row_ptr(&self) -> &[usize]

Row pointer slice (for external iteration).

pub fn targets(&self) -> &[NeuronId]

Target neuron slice.

pub fn weights(&self) -> &[f32]

Weight slice.

pub fn delays_slice(&self) -> &[DelayTicks]

Delay slice.

Trait Implementations

impl Clone for SynapticGraph

fn clone(&self) -> SynapticGraph

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for SynapticGraph

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

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for SynapticGraph

fn deserialize<D>(deserializer: D) -> Result<Self, D::Error>

where D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl PartialEq for SynapticGraph

fn eq(&self, other: &SynapticGraph) -> bool

Equality operator ==.

fn ne(&self, other: &Rhs) -> bool

Inequality operator !=.

impl Serialize for SynapticGraph

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

where __S: Serializer,

Serialize this value into the given Serde serializer.

impl StructuralPartialEq for SynapticGraph