LLMChainExtractor

class langchain.retrievers.document_compressors.chain_extract.LLMChainExtractor

Bases: BaseDocumentCompressor

Document compressor that uses an LLM chain to extract the relevant parts of documents.

Create a new model by parsing and validating input data from keyword arguments.

Raises ValidationError if the input data cannot be parsed to form a valid model.

param get_input: Callable[[str, Document], dict] = <function default_get_input>

Callable for constructing the chain input from the query and a Document.

param llm_chain: LLMChain [Required]

LLM wrapper to use for compressing documents.

async acompress_documents(documents: Sequence[Document], query: str, callbacks: List[BaseCallbackHandler] | BaseCallbackManager | None = None) → Sequence[Document]

Compress page content of raw documents asynchronously.

Parameters:

• documents (Sequence[Document]) –

• query (str) –

• callbacks (List[BaseCallbackHandler] | BaseCallbackManager | None) –

Return type:

Sequence[Document]

compress_documents(documents: Sequence[Document], query: str, callbacks: List[BaseCallbackHandler] | BaseCallbackManager | None = None) → Sequence[Document]

Compress page content of raw documents.

Parameters:

• documents (Sequence[Document]) –

• query (str) –

• callbacks (List[BaseCallbackHandler] | BaseCallbackManager | None) –

Return type:

Sequence[Document]

classmethod from_llm(llm: BaseLanguageModel, prompt: PromptTemplate | None = None, get_input: Callable[[str, Document], str] | None = None, llm_chain_kwargs: dict | None = None) → LLMChainExtractor

Initialize from LLM.

Parameters:

• llm (BaseLanguageModel) –

• prompt (PromptTemplate | None) –

• get_input (Callable[[str, Document], str] | None) –

• llm_chain_kwargs (dict | None) –

Return type:

LLMChainExtractor

Examples using LLMChainExtractor

• How to do retrieval with contextual compression