[Feature Request] Add buffermemory

[Ananderz]

I tried to have it elaborate on a previous question I asked and realized that it's missing a buffermemory. I am guessing it is using RetrieverQandA. I wonder if using ConversationalRetriverQandA with conversationalbuffermemory could be added to the solution or if there is a better way to add a memory of previous prompts and answers

[csunny]

I guess that the issue you are experiencing is related to the number of turns in the conversation history. Currently, it is set to 0 by default in our code. If you want to keep more history, you can modify the source code. In subsequent versions, we will make this a configurable option.

chat_retention_rounds: int = 0 

Source Code

class BaseChat(ABC):
    """DB-GPT Chat Service Base Module
    Include:
    stream_call():scene + prompt -> stream response
    nostream_call():scene + prompt -> nostream response
    """
    chat_scene: str = None
    llm_model: Any = None
    # By default, keep the last two rounds of conversation records as the context
    chat_retention_rounds: int = 0 

    class Config:
        """Configuration for this pydantic object."""

        arbitrary_types_allowed = True

[github-actions[bot]]

This issue has been marked as stale, because it has been over 30 days without any activity.

[github-actions[bot]]

This issue bas been closed, because it has been marked as stale and there has been no activity for over 7 days.