HuggingFaceEmbedding

Defined in: providers/huggingface/src/embedding.ts:34

Uses feature extraction from '@xenova/transformers' to generate embeddings. Per default the model XENOVA_ALL_MINILM_L6_V2 is used.

Can be changed by setting the modelType parameter in the constructor, e.g.:

new HuggingFaceEmbedding({
    modelType: HuggingFaceEmbeddingModelType.XENOVA_ALL_MPNET_BASE_V2,
});

Extends

• BaseEmbedding

HuggingFaceEmbedding<Options>(nodes, options?): Promise<BaseNode<Metadata>[]>

Defined in: providers/huggingface/src/embedding.ts:34

Uses feature extraction from '@xenova/transformers' to generate embeddings. Per default the model XENOVA_ALL_MINILM_L6_V2 is used.

Can be changed by setting the modelType parameter in the constructor, e.g.:

new HuggingFaceEmbedding({
    modelType: HuggingFaceEmbeddingModelType.XENOVA_ALL_MPNET_BASE_V2,
});

Type Parameters

• Options extends Record<string, unknown>

Parameters

nodes

BaseNode<Metadata>[]

options?

Options

Returns

Promise<BaseNode<Metadata>[]>

Constructors

new HuggingFaceEmbedding()

new HuggingFaceEmbedding(params): HuggingFaceEmbedding

Defined in: providers/huggingface/src/embedding.ts:42

Parameters

params

HuggingFaceEmbeddingParams = {}

Returns

HuggingFaceEmbedding

Overrides

BaseEmbedding.constructor

Properties

modelType

modelType: string = HuggingFaceEmbeddingModelType.XENOVA_ALL_MINILM_L6_V2

Defined in: providers/huggingface/src/embedding.ts:35

modelOptions

modelOptions: undefined | PretrainedModelOptions = {}

Defined in: providers/huggingface/src/embedding.ts:36

Methods

getExtractor()

getExtractor(): Promise<FeatureExtractionPipeline>

Defined in: providers/huggingface/src/embedding.ts:52

Returns

Promise<FeatureExtractionPipeline>

getTextEmbedding()

getTextEmbedding(text): Promise<number[]>

Defined in: providers/huggingface/src/embedding.ts:72

Parameters

text

string

Returns

Promise<number[]>

Overrides

BaseEmbedding.getTextEmbedding