Enum SeedSelectionStrategy 

pub enum SeedSelectionStrategy {
    StackedNSW,
    KSampledRandom {
        k: usize,
    },
}

Seed selection strategy for HNSW search initialization.

This controls how the search chooses its entrypoint(s). Different strategies can behave better on different datasets and scale regimes; treat this as a tuning knob and benchmark on your workload.

Variants

StackedNSW

Stacked NSW: Hierarchical multi-resolution graphs (default, best for large datasets) Uses entry point in highest layer, navigates down layer by layer.

KSampledRandom

K-Sampled Random Seeds: K random nodes per query (best for medium-scale 1M-25GB) Lower indexing overhead, but requires more samples on large datasets.

Fields

k: usize

Number of random seeds to sample (typically k or ef_search)

Trait Implementations

impl Clone for SeedSelectionStrategy

fn clone(&self) -> SeedSelectionStrategy

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for SeedSelectionStrategy

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

Formats the value using the given formatter.

impl Default for SeedSelectionStrategy

fn default() -> SeedSelectionStrategy

Returns the “default value” for a type.

impl PartialEq for SeedSelectionStrategy

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

Tests for self and other values to be equal, and is used by ==.

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

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl StructuralPartialEq for SeedSelectionStrategy