Method GetEmbeddings

Namespace

LMKit.Embeddings

Assembly

LM-Kit.NET.dll

GetEmbeddings(Attachment, CancellationToken)

Generates the embedding vector for a given file attachment. If the associated LM supports image embeddings, image attachments (for example, PNG, JPEG, TIFF) will be processed into embeddings; otherwise, text‑based attachments (for example, TXT, HTML) will be used.

public float[] GetEmbeddings(Attachment attachment, CancellationToken cancellationToken = default)

Parameters

attachment Attachment

The file attachment containing content, name, MIME type, and size. Supported formats include: PNG, BMP, GIF, PSD, PIC, JPEG, PNM, HDR, TGA, WEBP, TIFF, TXT, HTML.

cancellationToken CancellationToken

Optional. A CancellationToken for handling cancellation requests.

Returns

float[]

An array of floats representing the embedding vector of the attachment’s content. Each element contributes to the multidimensional representation of its semantic meaning.

Exceptions

ArgumentNullException

Thrown when attachment is null.

GetEmbeddings(ImageBuffer, CancellationToken)

Generates the embedding vector for a given image.

public float[] GetEmbeddings(ImageBuffer image, CancellationToken cancellationToken = default)

Parameters

image ImageBuffer

The image buffer to be converted into an embedding vector.

cancellationToken CancellationToken

Optional. A CancellationToken for handling cancellation requests.

Returns

float[]

An array of floats representing the embedding vector of the image. Each element contributes to the multidimensional representation of its visual content.

Exceptions

ArgumentNullException

Thrown when image is null.

InvalidModelException

Thrown when the current model does not support image embeddings.

GetEmbeddings(IEnumerable<ImageBuffer>, CancellationToken)

Generates embedding vectors for a collection of images.

public float[][] GetEmbeddings(IEnumerable<ImageBuffer> images, CancellationToken cancellationToken = default)

Parameters

images IEnumerable<ImageBuffer>

An enumerable of image buffers to be converted into embedding vectors.

cancellationToken CancellationToken

Optional. A CancellationToken for handling cancellation requests.

Returns

float[][]

An array of float arrays, each representing the embedding vector of the corresponding input image.

Exceptions

ArgumentNullException

Thrown when images is null or empty.

InvalidModelException

Thrown when the current model does not support image embeddings.

GetEmbeddings(string, CancellationToken)

Generates the embedding vector for a given text string. This vector represents the text in a high-dimensional space, enabling various natural language processing tasks by capturing semantic meaning.

public float[] GetEmbeddings(string text, CancellationToken cancellationToken = default)

Parameters

text string

The text string to be converted into an embedding vector. Cannot be null or empty.

cancellationToken CancellationToken

Optional. A CancellationToken for handling cancellation requests.

Returns

float[]

An array of floats representing the embedding vector of the input text. Each element contributes to the multidimensional representation of the text's semantic meaning.

Examples

using LMKit.Model;
using LMKit.Embeddings;
using System;

LM model = LM.LoadFromModelID("embeddinggemma-300m");
Embedder embedder = new Embedder(model);

// Generate embedding for text
float[] embedding = embedder.GetEmbeddings("Artificial intelligence is transforming industries.");

Console.WriteLine($"Embedding dimension: {embedding.Length}");
Console.WriteLine($"Embedding norm: {Math.Sqrt(embedding.Sum(x => x * x)):F4}");

Remarks

The dimension of the array is specified by the EmbeddingSize property.

Exceptions

ArgumentNullException

Thrown when the input text is null or empty, as an embedding vector cannot be generated from an empty input.

GetEmbeddings(IEnumerable<string>, CancellationToken)

Generates embedding vectors for a collection of text strings. Each vector represents a text in a high-dimensional space, enabling various natural language processing tasks by capturing semantic meaning.

public float[][] GetEmbeddings(IEnumerable<string> text, CancellationToken cancellationToken = default)

Parameters

text IEnumerable<string>

An enumerable of text strings to be converted into embedding vectors. Cannot be null or empty.

cancellationToken CancellationToken

Optional. A CancellationToken for handling cancellation requests.

Returns

float[][]

An array of float arrays, each representing the embedding vector of the corresponding input text.

Exceptions

ArgumentNullException

Thrown when the input text is null or empty.

GetEmbeddings(IList<int>, CancellationToken)

Generates the embedding vector for a given tokenized text. This vector represents the text in a high-dimensional space, enabling various natural language processing tasks by capturing semantic meaning.

public float[] GetEmbeddings(IList<int> tokens, CancellationToken cancellationToken = default)

Parameters

tokens IList<int>

The list of tokens to be converted into an embedding vector. Cannot be null or empty.

cancellationToken CancellationToken

Optional. A CancellationToken for handling cancellation requests.

Returns

float[]

An array of floats representing the embedding vector of the input tokens.

Remarks

The dimension of the array is specified by the EmbeddingSize property.

Exceptions

ArgumentNullException

Thrown when the input tokens are null or empty.

GetEmbeddings(IEnumerable<IList<int>>, CancellationToken)

Generates embedding vectors for a collection of tokenized texts. Each vector represents a text in a high-dimensional space, enabling various natural language processing tasks by capturing semantic meaning.

public float[][] GetEmbeddings(IEnumerable<IList<int>> tokens, CancellationToken cancellationToken = default)

Parameters

tokens IEnumerable<IList<int>>

An enumerable of token lists to be converted into embedding vectors. Cannot be null or empty.

cancellationToken CancellationToken

Optional. A CancellationToken for handling cancellation requests.

Returns

float[][]

An array of float arrays, each representing the embedding vector of the corresponding input tokens.

Exceptions

ArgumentNullException

Thrown when the input tokens are null or empty.