VectorDBQA Class — langchain Architecture

Architecture documentation for the VectorDBQA class in base.py from the langchain codebase.

Class python

Entity Profile

Dependency Diagram

graph TD
  ad09c90b_2890_30a7_1895_8c3576656b96["VectorDBQA"]
  cf809fc8_9926_bc74_caaa_077dffa64d09["BaseRetrievalQA"]
  ad09c90b_2890_30a7_1895_8c3576656b96 -->|extends| cf809fc8_9926_bc74_caaa_077dffa64d09
  d3aa4510_51a9_c393_bb4a_23fe112c52cd["base.py"]
  ad09c90b_2890_30a7_1895_8c3576656b96 -->|defined in| d3aa4510_51a9_c393_bb4a_23fe112c52cd
  ab594ff9_7823_bb7e_3f62_9bd5f568c3a7["validate_search_type()"]
  ad09c90b_2890_30a7_1895_8c3576656b96 -->|method| ab594ff9_7823_bb7e_3f62_9bd5f568c3a7
  55d39624_e191_dafc_2c96_8e7034fcaa32["_get_docs()"]
  ad09c90b_2890_30a7_1895_8c3576656b96 -->|method| 55d39624_e191_dafc_2c96_8e7034fcaa32
  5cda0264_b1c1_3e8e_120d_b7697af0e3df["_aget_docs()"]
  ad09c90b_2890_30a7_1895_8c3576656b96 -->|method| 5cda0264_b1c1_3e8e_120d_b7697af0e3df
  1daea205_6ebc_225b_7865_d9eceb158316["_chain_type()"]
  ad09c90b_2890_30a7_1895_8c3576656b96 -->|method| 1daea205_6ebc_225b_7865_d9eceb158316

Relationship Graph

Source Code

libs/langchain/langchain_classic/chains/retrieval_qa/base.py lines 307–368

class VectorDBQA(BaseRetrievalQA):
    """Chain for question-answering against a vector database."""

    vectorstore: VectorStore = Field(exclude=True, alias="vectorstore")
    """Vector Database to connect to."""
    k: int = 4
    """Number of documents to query for."""
    search_type: str = "similarity"
    """Search type to use over vectorstore. `similarity` or `mmr`."""
    search_kwargs: dict[str, Any] = Field(default_factory=dict)
    """Extra search args."""

    @model_validator(mode="before")
    @classmethod
    def validate_search_type(cls, values: dict) -> Any:
        """Validate search type."""
        if "search_type" in values:
            search_type = values["search_type"]
            if search_type not in ("similarity", "mmr"):
                msg = f"search_type of {search_type} not allowed."
                raise ValueError(msg)
        return values

    @override
    def _get_docs(
        self,
        question: str,
        *,
        run_manager: CallbackManagerForChainRun,
    ) -> list[Document]:
        """Get docs."""
        if self.search_type == "similarity":
            docs = self.vectorstore.similarity_search(
                question,
                k=self.k,
                **self.search_kwargs,
            )
        elif self.search_type == "mmr":
            docs = self.vectorstore.max_marginal_relevance_search(
                question,
                k=self.k,
                **self.search_kwargs,
            )
        else:
            msg = f"search_type of {self.search_type} not allowed."
            raise ValueError(msg)
        return docs

    async def _aget_docs(
        self,
        question: str,
        *,
        run_manager: AsyncCallbackManagerForChainRun,
    ) -> list[Document]:
        """Get docs."""
        msg = "VectorDBQA does not support async"
        raise NotImplementedError(msg)

    @property
    def _chain_type(self) -> str:
        """Return the chain type."""
        return "vector_db_qa"