QAGenerationChain Class — langchain Architecture

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

Class python

Entity Profile

Dependency Diagram

graph TD
  2ce04a9b_5637_00e2_15a2_00f9ad51f684["QAGenerationChain"]
  f3cef70e_11b0_61c9_7ec0_7308f4b45056["Chain"]
  2ce04a9b_5637_00e2_15a2_00f9ad51f684 -->|extends| f3cef70e_11b0_61c9_7ec0_7308f4b45056
  abcb2fe7_a066_23d4_b7f5_15af8ffd2920["base.py"]
  2ce04a9b_5637_00e2_15a2_00f9ad51f684 -->|defined in| abcb2fe7_a066_23d4_b7f5_15af8ffd2920
  610e495a_2355_f81c_2c3f_0bfb090f206d["from_llm()"]
  2ce04a9b_5637_00e2_15a2_00f9ad51f684 -->|method| 610e495a_2355_f81c_2c3f_0bfb090f206d
  b924f598_539c_def7_0504_069160fb0d62["_chain_type()"]
  2ce04a9b_5637_00e2_15a2_00f9ad51f684 -->|method| b924f598_539c_def7_0504_069160fb0d62
  045d06b7_df8e_4a58_b958_b4e00a039900["input_keys()"]
  2ce04a9b_5637_00e2_15a2_00f9ad51f684 -->|method| 045d06b7_df8e_4a58_b958_b4e00a039900
  17bb8513_b3d0_a798_7e2f_d9eaa8f8ad95["output_keys()"]
  2ce04a9b_5637_00e2_15a2_00f9ad51f684 -->|method| 17bb8513_b3d0_a798_7e2f_d9eaa8f8ad95
  62e4d560_1563_c959_ce77_10d90ce81c1f["_call()"]
  2ce04a9b_5637_00e2_15a2_00f9ad51f684 -->|method| 62e4d560_1563_c959_ce77_10d90ce81c1f

Relationship Graph

Source Code

libs/langchain/langchain_classic/chains/qa_generation/base.py lines 27–127

class QAGenerationChain(Chain):
    """Base class for question-answer generation chains.

    This class is deprecated. See below for an alternative implementation.

    Advantages of this implementation include:

    - Supports async and streaming;
    - Surfaces prompt and text splitter for easier customization;
    - Use of JsonOutputParser supports JSONPatch operations in streaming mode,
        as well as robustness to markdown.

        ```python
        from langchain_classic.chains.qa_generation.prompt import (
            CHAT_PROMPT as prompt,
        )

        # Note: import PROMPT if using a legacy non-chat model.
        from langchain_core.output_parsers import JsonOutputParser
        from langchain_core.runnables import (
            RunnableLambda,
            RunnableParallel,
            RunnablePassthrough,
        )
        from langchain_core.runnables.base import RunnableEach
        from langchain_openai import ChatOpenAI
        from langchain_text_splitters import RecursiveCharacterTextSplitter

        model = ChatOpenAI()
        text_splitter = RecursiveCharacterTextSplitter(chunk_overlap=500)
        split_text = RunnableLambda(lambda x: text_splitter.create_documents([x]))

        chain = RunnableParallel(
            text=RunnablePassthrough(),
            questions=(
                split_text | RunnableEach(bound=prompt | model | JsonOutputParser())
            ),
        )
        ```
    """

    llm_chain: LLMChain
    """LLM Chain that generates responses from user input and context."""
    text_splitter: TextSplitter = Field(
        default=RecursiveCharacterTextSplitter(chunk_overlap=500),
    )
    """Text splitter that splits the input into chunks."""
    input_key: str = "text"
    """Key of the input to the chain."""
    output_key: str = "questions"
    """Key of the output of the chain."""
    k: int | None = None
    """Number of questions to generate."""

    @classmethod
    def from_llm(
        cls,
        llm: BaseLanguageModel,
        prompt: BasePromptTemplate | None = None,
        **kwargs: Any,
    ) -> QAGenerationChain:
        """Create a QAGenerationChain from a language model.

        Args:
            llm: a language model
            prompt: a prompt template
            **kwargs: additional arguments

        Returns:
            a QAGenerationChain class
        """
        _prompt = prompt or PROMPT_SELECTOR.get_prompt(llm)
        chain = LLMChain(llm=llm, prompt=_prompt)
        return cls(llm_chain=chain, **kwargs)

    @property
    def _chain_type(self) -> str:
        raise NotImplementedError

    @property
    @override