Class LocalEmbeddingProvider

Namespace

Mythosia.AI.Rag.Embeddings

Assembly

Mythosia.AI.Rag.dll

Local embedding provider using feature hashing (hashing trick). Works without any external API key. Suitable for prototyping and small-scale use. Quality is lower than neural embeddings (OpenAI, etc.) but requires zero configuration.

public class LocalEmbeddingProvider : IEmbeddingProvider

Inheritance

object

LocalEmbeddingProvider

Implements

IEmbeddingProvider

Inherited Members

object.Equals(object)

object.Equals(object, object)

object.GetHashCode()

object.GetType()

object.MemberwiseClone()

object.ReferenceEquals(object, object)

object.ToString()

Constructors

LocalEmbeddingProvider(int)

Creates a local embedding provider with the specified vector dimensionality. Higher dimensions capture more information but use more memory.

public LocalEmbeddingProvider(int dimensions = 1024)

Parameters

dimensions int

Vector dimensionality. Default is 1024.

Properties

Dimensions

The dimensionality of the embedding vectors produced by this provider.

public int Dimensions { get; }

Property Value

int

Methods

GetEmbeddingAsync(string, CancellationToken)

Generates an embedding vector for a single text input.

public Task<float[]> GetEmbeddingAsync(string text, CancellationToken cancellationToken = default)

Parameters

text string

cancellationToken CancellationToken

Returns

Task<float[]>

GetEmbeddingsAsync(IEnumerable<string>, CancellationToken)

Generates embedding vectors for multiple text inputs in a single batch call.

public Task<IReadOnlyList<float[]>> GetEmbeddingsAsync(IEnumerable<string> texts, CancellationToken cancellationToken = default)

Parameters

texts IEnumerable<string>

cancellationToken CancellationToken

Returns

Task<IReadOnlyList<float[]>>