SparseVector

Represents a sparse vector using COO (Coordinate) format: a set of (index, value) pairs with non-negative, ascending indices.

public readonly struct SparseVector<T> : IEquatable<SparseVector<T>>
{
    public SparseVector(ReadOnlyMemory<int> indices, ReadOnlyMemory<T> values) { ... }
    public int Count { get; }
    public ReadOnlyMemory<int> Indices { get; }
    public ReadOnlyMemory<T> Values { get; }
}

PARAMETERS:

  • Count (int) -

    The number of non-zero elements.

  • Indices (ReadOnlyMemory) -

    The indices of non-zero elements, in ascending order.

  • Values (ReadOnlyMemory) -

    The values of non-zero elements, in the same order as Indices.

METHODS:

  • SparseVector(ReadOnlyMemory<int> indices, ReadOnlyMemory<T> values) (constructor) -

    Creates a sparse vector from parallel collections of indices and values. The indices must be non-negative and sorted in ascending order.

Example

using Milvus.Client.V2;
using Milvus.Client.V2.Requests.Dql;
using Milvus.Client.V2.Responses.Dql;
using Milvus.Client.V2.Types;

using var client = new MilvusClientV2(new ConnectConfig
{
    Uri = "localhost:19530",
    Username = "root",
    Password = "Milvus"
});

var sparseVector = new SparseVector<float>(
    new int[] { 2, 5, 8 },
    new float[] { 0.3f, 0.1f, 0.7f });

SearchResp results = await client.SearchAsync(new SearchReq
{
    CollectionName = "my_collection",
    VectorFieldName = "sparse",
    SparseVectors = [sparseVector],
    MetricType = MetricType.Bm25,
    Limit = 5
});

Try Managed Milvus for Free

Zilliz Cloud is hassle-free, powered by Milvus and 10x faster.

Get Started
Feedback

Was this page helpful?