Package org.apache.lucene.util.hnsw

Class HnswGraphBuilder<T>

java.lang.Object

org.apache.lucene.util.hnsw.HnswGraphBuilder<T>

Type Parameters:

T - the type of vector

public final class HnswGraphBuilder<T>
extends Object

Builder for HNSW graph. See HnswGraph for a gloss on the algorithm and the meaning of the hyperparameters.

Field Summary

Fields

Modifier and Type	Field	Description
static String	HNSW_COMPONENT	A name for the HNSW component for the info-stream *
static long	randSeed	Random seed for level generation; public to expose for testing *

Method Summary

Modifier and Type	Method	Description
void	addGraphNode(int node, RandomAccessVectorValues<T> values)
void	addGraphNode(int node, T value)	Inserts a doc with vector value to the graph
OnHeapHnswGraph	build(RandomAccessVectorValues<T> vectorsToAdd)	Reads all the vectors from two copies of a RandomAccessVectorValues.
static <T> HnswGraphBuilder<T>	create(RandomAccessVectorValues<T> vectors, VectorEncoding vectorEncoding, VectorSimilarityFunction similarityFunction, int M, int beamWidth, long seed)
static <T> HnswGraphBuilder<T>	create(RandomAccessVectorValues<T> vectors, VectorEncoding vectorEncoding, VectorSimilarityFunction similarityFunction, int M, int beamWidth, long seed, HnswGraph initializerGraph, Map<Integer,Integer> oldToNewOrdinalMap)
OnHeapHnswGraph	getGraph()
void	setInfoStream(InfoStream infoStream)	Set info-stream to output debugging information *

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Detail

HNSW_COMPONENT

public static final String HNSW_COMPONENT

A name for the HNSW component for the info-stream *

See Also:

Constant Field Values

randSeed

public static long randSeed

Random seed for level generation; public to expose for testing *

Method Detail

create

public static <T> HnswGraphBuilder<T> create(RandomAccessVectorValues<T> vectors,
                                             VectorEncoding vectorEncoding,
                                             VectorSimilarityFunction similarityFunction,
                                             int M,
                                             int beamWidth,
                                             long seed)
                                      throws IOException

Throws:

IOException

create

public static <T> HnswGraphBuilder<T> create(RandomAccessVectorValues<T> vectors,
                                             VectorEncoding vectorEncoding,
                                             VectorSimilarityFunction similarityFunction,
                                             int M,
                                             int beamWidth,
                                             long seed,
                                             HnswGraph initializerGraph,
                                             Map<Integer,Integer> oldToNewOrdinalMap)
                                      throws IOException

Throws:

IOException

build

public OnHeapHnswGraph build(RandomAccessVectorValues<T> vectorsToAdd)
                      throws IOException

Reads all the vectors from two copies of a RandomAccessVectorValues. Providing two copies enables efficient retrieval without extra data copying, while avoiding collision of the returned values.

Parameters:

vectorsToAdd - the vectors for which to build a nearest neighbors graph. Must be an independent accessor for the vectors

Throws:

IOException

setInfoStream

public void setInfoStream(InfoStream infoStream)

Set info-stream to output debugging information *

getGraph

public OnHeapHnswGraph getGraph()

addGraphNode

public void addGraphNode(int node,
                         T value)
                  throws IOException

Inserts a doc with vector value to the graph

Throws:

IOException

addGraphNode

public void addGraphNode(int node,
                         RandomAccessVectorValues<T> values)
                  throws IOException

Throws:

IOException