Hugging Face

HuggingFaceEmbedding #

Bases: BaseEmbedding

HuggingFace sentence_transformers embedding models.

Attributes:

Name	Type	Description
`model_name`	`str`	Hugging Face model to be used. Defaults to `sentence-transformers/all-MiniLM-L6-v2`.
`device`	`str`	Device to run the model on. Supports `cpu` and `cuda`. Defaults to `cpu`.

Example

from beekeeper.embeddings.huggingface import HuggingFaceEmbedding

embedding = HuggingFaceEmbedding()

embed_documents(documents: list[Document]) -> list[Document]

Embed a list of documents and assign the computed embeddings to the 'embedding' attribute.

Parameters:

Name	Type	Description	Default
`documents`	`list[Document]`	List of documents to compute embeddings.	required

embed_text(input: str | list[str]) -> list[Embedding]

Embed one or more text strings.

Parameters:

Name	Type	Description	Default
`input`	`list[str]`	Input for which to compute embeddings.	required