Module org.apache.lucene.core
Package org.apache.lucene.util.quantization
package org.apache.lucene.util.quantization
Provides quantization methods for scaling vector values to smaller data types and possibly fewer
dimensions
-
ClassDescriptionA version of
ByteVectorValues
, but additionally retrieving score correction offset for Scalar quantization scores.Quantized vector readerRandom access values forbyte[]
, but also includes accessing the score correction constant for the current vector in the buffer.Calculates and adjust the scores correctly for quantized vectors given the scalar quantization parametersCompares two byte vectorsCalculates dot product on quantized vectors, applying the appropriate correctionsCalculates euclidean distance on quantized vectors, applying the appropriate correctionsCalculates max inner product on quantized vectors, applying the appropriate correctionsWill scalar quantize float vectors into `int8` byte values.This class is used to correlate the scores of the nearest neighbors with the errors in the scores.