Class: HuggingFaceEmbedding

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,
});

Hierarchy

• BaseEmbedding

  ↳ HuggingFaceEmbedding

Constructors

constructor

• new HuggingFaceEmbedding(init?)

Parameters

Name	Type
init?	Partial<HuggingFaceEmbedding>

Overrides

BaseEmbedding.constructor

Defined in

packages/core/src/embeddings/HuggingFaceEmbedding.ts:26

Properties

extractor

• Private extractor: any

Defined in

packages/core/src/embeddings/HuggingFaceEmbedding.ts:24

---

modelType

• modelType: string = HuggingFaceEmbeddingModelType.XENOVA_ALL_MINILM_L6_V2

Defined in

packages/core/src/embeddings/HuggingFaceEmbedding.ts:22

Methods

getExtractor

▸ getExtractor(): Promise<any>

Returns

Promise<any>

Defined in

packages/core/src/embeddings/HuggingFaceEmbedding.ts:31

---

getQueryEmbedding

▸ getQueryEmbedding(query): Promise<number[]>

Parameters

Name	Type
query	string

Returns

Promise<number[]>

Overrides

BaseEmbedding.getQueryEmbedding

Defined in

packages/core/src/embeddings/HuggingFaceEmbedding.ts:45

---

getTextEmbedding

▸ getTextEmbedding(text): Promise<number[]>

Parameters

Name	Type
text	string

Returns

Promise<number[]>

Overrides

BaseEmbedding.getTextEmbedding

Defined in

packages/core/src/embeddings/HuggingFaceEmbedding.ts:39

---

similarity

▸ similarity(embedding1, embedding2, mode?): number

Parameters

Name	Type	Default value
embedding1	number[]	undefined
embedding2	number[]	undefined
mode	SimilarityType	SimilarityType.DEFAULT

Returns

number

Inherited from

BaseEmbedding.similarity

Defined in

packages/core/src/embeddings/types.ts:14