embeddings-base/ai.koog.embeddings.base/Vector

Vector

@Serializable

data class Vector(val values: List<Double>)(source)

Represents a vector of floating-point values. Used for storing embeddings of text.

Constructors

Vector

constructor(values: List<Double>)

Properties

dimension

val dimension: Int

Returns the dimension (size) of the vector.

values

val values: List<Double>

The floating-point values that make up the vector.

Functions

cosineSimilarity

fun cosineSimilarity(other: Vector): Double

Calculates the cosine similarity between this vector and another vector. The result is a value between -1 and 1, where 1 means the vectors are identical, 0 means they are orthogonal, and -1 means they are completely opposite.

dotProduct

infix fun dotProduct(other: Vector): Double

euclideanDistance

fun euclideanDistance(other: Vector): Double

Calculates the Euclidean distance between this vector and another vector. The result is a non-negative value, where 0 means the vectors are identical.

isNull

fun isNull(): Boolean

magnitude

fun magnitude(): Double