QuestionGeneratorChain Class — langchain Architecture

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

Class python

Entity Profile

Dependency Diagram

graph TD
  c67ceace_d96a_acd7_994d_55c9d0939c8f["QuestionGeneratorChain"]
  ccf50fe1_4990_cf19_1e2d_25efe83f53c5["LLMChain"]
  c67ceace_d96a_acd7_994d_55c9d0939c8f -->|extends| ccf50fe1_4990_cf19_1e2d_25efe83f53c5
  4c01b940_b202_29af_2ff5_d0dfa5e5e804["base.py"]
  c67ceace_d96a_acd7_994d_55c9d0939c8f -->|defined in| 4c01b940_b202_29af_2ff5_d0dfa5e5e804
  6a4daf80_4124_cad2_7781_e4567c89432f["is_lc_serializable()"]
  c67ceace_d96a_acd7_994d_55c9d0939c8f -->|method| 6a4daf80_4124_cad2_7781_e4567c89432f
  8c50a119_1cd3_4545_73b7_ce942f876b6a["input_keys()"]
  c67ceace_d96a_acd7_994d_55c9d0939c8f -->|method| 8c50a119_1cd3_4545_73b7_ce942f876b6a

Relationship Graph

Source Code

libs/langchain/langchain_classic/chains/flare/base.py lines 41–55

class QuestionGeneratorChain(LLMChain):
    """Chain that generates questions from uncertain spans."""

    prompt: BasePromptTemplate = QUESTION_GENERATOR_PROMPT
    """Prompt template for the chain."""

    @classmethod
    @override
    def is_lc_serializable(cls) -> bool:
        return False

    @property
    def input_keys(self) -> list[str]:
        """Input keys for the chain."""
        return ["user_input", "context", "response"]