from_messages() — langchain Function Reference

Architecture documentation for the from_messages() function in summary.py from the langchain codebase.

Function python LangChainCore ApiManagement calls 1

Entity Profile

LangChainCore→ ApiManagement→ from_messages() — langchain Function Reference

Dependency Diagram

graph TD
  1f012672_a4f9_fcdb_ebce_4907d85c70cf["from_messages()"]
  896d889a_d228_6c16_6cdd_5799f6e5f307["ConversationSummaryMemory"]
  1f012672_a4f9_fcdb_ebce_4907d85c70cf -->|defined in| 896d889a_d228_6c16_6cdd_5799f6e5f307
  440abe33_20df_ee02_d771_d5bb5d39241f["predict_new_summary()"]
  1f012672_a4f9_fcdb_ebce_4907d85c70cf -->|calls| 440abe33_20df_ee02_d771_d5bb5d39241f
  style 1f012672_a4f9_fcdb_ebce_4907d85c70cf fill:#6366f1,stroke:#818cf8,color:#fff

Relationship Graph

Source Code

libs/langchain/langchain_classic/memory/summary.py lines 103–128

    def from_messages(
        cls,
        llm: BaseLanguageModel,
        chat_memory: BaseChatMessageHistory,
        *,
        summarize_step: int = 2,
        **kwargs: Any,
    ) -> ConversationSummaryMemory:
        """Create a ConversationSummaryMemory from a list of messages.

        Args:
            llm: The language model to use for summarization.
            chat_memory: The chat history to summarize.
            summarize_step: Number of messages to summarize at a time.
            **kwargs: Additional keyword arguments to pass to the class.

        Returns:
            An instance of ConversationSummaryMemory with the summarized history.
        """
        obj = cls(llm=llm, chat_memory=chat_memory, **kwargs)
        for i in range(0, len(obj.chat_memory.messages), summarize_step):
            obj.buffer = obj.predict_new_summary(
                obj.chat_memory.messages[i : i + summarize_step],
                obj.buffer,
            )
        return obj