Struct InMemoryVamana 

pub struct InMemoryVamana<T: Scalar + Send + Sync> { /* private fields */ }

Implementations

impl<T: Scalar + Send + Sync> InMemoryVamana<T>

pub fn new( adj_list: DashMap<Id, Vec<Id>>, id_to_point: DashMap<Id, Array1<T>>, ) -> Self

pub fn graph(&self) -> &DashMap<Id, Vec<Id>>

This can be useful for serializing to disk. Usually the points are already stored elsewhere, so serializing the graph alone can save space. To deserialize, collect the points, deserialize this graph, and use InMemoryVamana::new. This can also be used to introspect the graph, e.g. for debugging, analysis, or research.

pub fn init_random_index( dataset: Vec<(Id, Array1<T>)>, metric: Metric<T>, params: VamanaParams, medoid_sample_size: usize, ) -> Vamana<T, Self>

pub fn build_index( dataset: Vec<(Id, Array1<T>)>, metric: Metric<T>, params: VamanaParams, medoid_sample_size: usize, ) -> Vamana<T, Self>

Trait Implementations

impl<T: Clone + Scalar + Send + Sync> Clone for InMemoryVamana<T>

fn clone(&self) -> InMemoryVamana<T>

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl<T: Default + Scalar + Send + Sync> Default for InMemoryVamana<T>

fn default() -> InMemoryVamana<T>

Returns the “default value” for a type.

impl<T: Scalar + Send + Sync> GreedySearchable<T> for InMemoryVamana<T>

fn get_point(&self, id: Id) -> Array1<T>

fn get_out_neighbors(&self, id: Id) -> Vec<Id>

impl<T: Scalar + Send + Sync> VamanaDatastore<T> for InMemoryVamana<T>

fn set_point(&self, id: Id, point: Array1<T>)

fn set_out_neighbors(&self, id: Id, neighbors: Vec<Id>)