Method SearchSimilarVectorsAsync

Namespace

LMKit.Data.Storage.Qdrant

Assembly

LM-Kit.NET.Data.Connectors.Qdrant.dll

SearchSimilarVectorsAsync(string, float[], uint, bool, bool, CancellationToken)

Asynchronously searches for vectors within the specified collection that are similar to a given query vector.

public Task<List<(PointEntry Point, float Score)>> SearchSimilarVectorsAsync(string collectionIdentifier, float[] vector, uint limit, bool getVector, bool getMetadata, CancellationToken cancellationToken = default)

Parameters

collectionIdentifier string

The name of the collection to search within.

vector float[]

The query vector used to compute similarity against stored vectors. The length of this vector should match the collection's defined vector size.

limit uint

The maximum number of similar vector entries to return.

getVector bool

If true, the vector data for each matching entry will be included in the results.

getMetadata bool

If true, the metadata for each matching entry will be included in the results.

cancellationToken CancellationToken

An optional token to cancel the asynchronous search operation.

Returns

Task<List<(PointEntry Point, float Score)>>

A task representing the asynchronous search operation. The task result is a list of tuples, where each tuple contains a PointEntry and a float similarity score. The similarity score is computed using the underlying vector similarity algorithm defined by the implementation.