ScoreMetric

Cosine similarity between two vectors, ranging from -1 to 1. A value of 1 indicates identical direction, 0 indicates orthogonality, and -1 indicates opposite direction. Higher values mean greater similarity.

Cosine distance between two vectors, defined as 1 - cosine_similarity, ranging from 0 to 2. A value of 0 indicates identical vectors, while 2 indicates opposite vectors. Lower values mean greater similarity.

Dot product (inner product) of two vectors. The value is unbounded and depends on vector magnitudes. Higher values indicate greater similarity when vectors are normalized.

Euclidean (L2) distance between two vectors, ranging from 0 to infinity. A value of 0 indicates identical vectors. Lower values mean greater similarity.

BM25 (Best Matching 25) text relevance score, commonly used in full-text search. The value is non-negative and unbounded. Higher values indicate greater relevance.

A hybrid score combining multiple retrieval strategies (e.g., vector similarity and keyword search). The range and interpretation depend on the specific combination and normalization used by the storage implementation.

A custom scoring metric defined by the storage implementation. The range and interpretation are implementation-specific.

Returns a representation of an immutable list of all enum entries, in the order they're declared.

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.)

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