_FakeTrajectoryChatModel Class — langchain Architecture

Architecture documentation for the _FakeTrajectoryChatModel class in test_eval_chain.py from the langchain codebase.

Class python

Entity Profile

Dependency Diagram

graph TD
  b0bd79d7_7c19_7abd_ea39_9fe2eaf38839["_FakeTrajectoryChatModel"]
  08f149d6_0ef5_52a4_96df_fceeb68e4ea8["FakeChatModel"]
  b0bd79d7_7c19_7abd_ea39_9fe2eaf38839 -->|extends| 08f149d6_0ef5_52a4_96df_fceeb68e4ea8
  1ff39fbd_e89b_6287_757b_009d26497b03["test_eval_chain.py"]
  b0bd79d7_7c19_7abd_ea39_9fe2eaf38839 -->|defined in| 1ff39fbd_e89b_6287_757b_009d26497b03
  1f5b87f2_3a5c_ad37_a5fd_1a4aa2aab482["_call()"]
  b0bd79d7_7c19_7abd_ea39_9fe2eaf38839 -->|method| 1f5b87f2_3a5c_ad37_a5fd_1a4aa2aab482

Relationship Graph

Source Code

libs/langchain/tests/unit_tests/evaluation/agents/test_eval_chain.py lines 42–60

class _FakeTrajectoryChatModel(FakeChatModel):
    queries: dict = Field(default_factory=dict)
    sequential_responses: bool | None = False
    response_index: int = 0

    @override
    def _call(
        self,
        messages: list[BaseMessage],
        stop: list[str] | None = None,
        run_manager: CallbackManagerForLLMRun | None = None,
        **kwargs: Any,
    ) -> str:
        if self.sequential_responses:
            response = self.queries[list(self.queries.keys())[self.response_index]]
            self.response_index = self.response_index + 1
            return response
        prompt = messages[0].content
        return self.queries[prompt]