Package org.hibernate.search.engine.backend.types

Enum VectorSimilarity

java.lang.Object

java.lang.Enum<VectorSimilarity>

org.hibernate.search.engine.backend.types.VectorSimilarity

All Implemented Interfaces:

Serializable, Comparable<VectorSimilarity>

@Incubating
public enum VectorSimilarity
extends Enum<VectorSimilarity>

Defines a function to calculate the vector similarity, i.e. distance between two vectors.

Note, some backends or their distributions may not support all of available similarity options.

Enum Constant Summary

Enum Constants

Enum Constant	Description
COSINE	Cosine similarity.
DEFAULT	Use the backend-specific default.
DOT_PRODUCT	Inner product (dot product in particular).
L2	L2 (Euclidean) norm.
MAX_INNER_PRODUCT	Similar to DOT_PRODUCT but does not require vector normalization.

Method Summary

Modifier and Type	Method	Description
static VectorSimilarity	valueOf(String name)	Returns the enum constant of this type with the specified name.
static VectorSimilarity[]	values()	Returns an array containing the constants of this enum type, in the order they are declared.

Methods inherited from class java.lang.Enum

clone, compareTo, equals, finalize, getDeclaringClass, hashCode, name, ordinal, toString, valueOf

Methods inherited from class java.lang.Object

getClass, notify, notifyAll, wait, wait, wait

Enum Constant Details

DEFAULT

public static final VectorSimilarity DEFAULT

Use the backend-specific default.

L2

public static final VectorSimilarity L2

L2 (Euclidean) norm. Distance is calculated as d(x,y) = \sum_(i=1) ^(n) (x_i - y_i)^2 and similarity function is s = 1 / (1+d)

DOT_PRODUCT

public static final VectorSimilarity DOT_PRODUCT

Inner product (dot product in particular). Distance is calculated as d(x,y) = \sum_(i=1) ^(n) x_i*y_i , similarity function is s = 1 / (1+d) , but may differ between backends.

WARNING:: to use this similarity efficiently, both index and search vectors must be normalized, otherwise search may produce poor results. Floating point vectors must be normalized to be of unit length, while byte vectors should simply all have the same norm.

COSINE

public static final VectorSimilarity COSINE

Cosine similarity. Distance is calculated as d(x,y) = 1 - \sum_(i=1) ^(n) x_i*y_i / ( \sqrt( \sum_(i=1) ^(n) x_i^2 ) \sqrt( \sum_(i=1) ^(n) y_i^2 ) , similarity function is s = 1 / (1+d) , but may differ between backends.

MAX_INNER_PRODUCT

public static final VectorSimilarity MAX_INNER_PRODUCT

Similar to DOT_PRODUCT but does not require vector normalization. Distance is calculated as d(x,y) = \sum_(i=1) ^(n) (x_i - y_i)^2 and similarity function is d < 0 ? s = 1 / (1 - d) : s = d + 1

Method Details

values

public static VectorSimilarity[] values()

Returns an array containing the constants of this enum type, in the order they are declared.

Returns:

an array containing the constants of this enum type, in the order they are declared

valueOf

public static VectorSimilarity valueOf(String name)

Returns the enum constant of this type with the specified name. The string must match exactly an identifier used to declare an enum constant in this type. (Extraneous whitespace characters are not permitted.)

Parameters:

name - the name of the enum constant to be returned.

Returns:

the enum constant with the specified name

Throws:

IllegalArgumentException - if this enum type has no constant with the specified name

NullPointerException - if the argument is null