Navigation

org.apache.lucene.util.quantization 5

C OptimizedScalarQuantizer

C QuantizedByteVectorValues

I QuantizedVectorsReader

I ScalarQuantizedVectorSimilarity

C ScalarQuantizer

ScalarQuantizedVectorSimilarity — members 2

M from_vector_similarity(sim, const_multiplier, bits) ScalarQuantizedVectorSimilarity

M score(query_vector, query_vector_offset, stored_vector, vector_offset) float

org.apache.lucene.util.quantization.ScalarQuantizedVectorSimilarity

interface abstract 2 members

Calculates and adjust the scores correctly for quantized vectors given the scalar quantization parameters

Class Methods

ScalarQuantizedVectorSimilarity from_vector_similarity ( VectorSimilarityFunction sim, float const_multiplier, byte bits )

Java: fromVectorSimilarity(org.apache.lucene.index.VectorSimilarityFunction sim, float constMultiplier, byte bits)

Creates a ScalarQuantizedVectorSimilarity from a VectorSimilarityFunction and the constant multiplier used for quantization.

name	type	description
sim	org.apache.lucene.index.VectorSimilarityFunction	similarity function
const_multiplier	float	constant multiplier used for quantization
bits	byte	number of bits used for quantization

Returns: a ScalarQuantizedVectorSimilarity that applies the appropriate corrections

float score ( byte[] query_vector, float query_vector_offset, byte[] stored_vector, float vector_offset )

Java: score(byte[] queryVector, float queryVectorOffset, byte[] storedVector, float vectorOffset)