Member Junction
    Preparing search index...

    Interface VectorClusterJob

    A clustering run over a store snapshot (optionally a subset of its rows).

    interface VectorClusterJob {
        Algorithm: VectorClusterAlgorithm;
        Candidates: Int32Array<ArrayBufferLike>;
        Epsilon?: number;
        K?: number;
        MaxIterations?: number;
        Metric: DistanceMetric;
        MinPoints?: number;
        Snapshot: VectorStoreSnapshot;
        Tolerance?: number;
    }
    Index

    Properties

    Algorithm Candidates Epsilon? K? MaxIterations? Metric MinPoints? Snapshot Tolerance?

    Properties

    Algorithm: VectorClusterAlgorithm
    Candidates: Int32Array<ArrayBufferLike>

    Ascending rows to cluster, or null for every live row

    Epsilon?: number

    DBSCAN: neighbourhood radius (0-1, exclusive)

    K?: number

    K-Means: number of clusters

    MaxIterations?: number

    K-Means: iteration cap

    Metric: DistanceMetric
    MinPoints?: number

    DBSCAN: minimum neighbours for a core point

    Snapshot: VectorStoreSnapshot
    Tolerance?: number

    K-Means: convergence tolerance

    Settings

    Member Visibility

    On This Page

    Properties