langchain.embeddings.huggingface.HuggingFaceInstructEmbeddings¶

class langchain.embeddings.huggingface.HuggingFaceInstructEmbeddings(*, client: Any = None, model_name: str = 'hkunlp/instructor-large', cache_folder: Optional[str] = None, model_kwargs: Dict[str, Any] = None, encode_kwargs: Dict[str, Any] = None, embed_instruction: str = 'Represent the document for retrieval: ', query_instruction: str = 'Represent the question for retrieving supporting documents: ')[source]¶

Bases: BaseModel, Embeddings

Wrapper around sentence_transformers embedding models.

To use, you should have the sentence_transformers and InstructorEmbedding python packages installed.

Example

from langchain.embeddings import HuggingFaceInstructEmbeddings

model_name = "hkunlp/instructor-large"
model_kwargs = {'device': 'cpu'}
encode_kwargs = {'normalize_embeddings': True}
hf = HuggingFaceInstructEmbeddings(
    model_name=model_name,
    model_kwargs=model_kwargs,
    encode_kwargs=encode_kwargs
)

Initialize the sentence_transformer.

param cache_folder: Optional[str] = None¶

Path to store models. Can be also set by SENTENCE_TRANSFORMERS_HOME environment variable.

param embed_instruction: str = 'Represent the document for retrieval: '¶

Instruction to use for embedding documents.

param encode_kwargs: Dict[str, Any] [Optional]¶

Key word arguments to pass when calling the encode method of the model.

param model_kwargs: Dict[str, Any] [Optional]¶

Key word arguments to pass to the model.

param model_name: str = 'hkunlp/instructor-large'¶

Model name to use.

param query_instruction: str = 'Represent the question for retrieving supporting documents: '¶

Instruction to use for embedding query.

embed_documents(texts: List[str]) → List[List[float]][source]¶

Compute doc embeddings using a HuggingFace instruct model.

Parameters

texts – The list of texts to embed.

Returns

List of embeddings, one for each text.

embed_query(text: str) → List[float][source]¶

Compute query embeddings using a HuggingFace instruct model.

Parameters

text – The text to embed.

Returns

Embeddings for the text.

model Config[source]¶

Bases: object

Configuration for this pydantic object.

extra = 'forbid'¶

Examples using HuggingFaceInstructEmbeddings¶