Langchain-Chatchat/models/fastchat_llm.py

from abc import ABC
import requests
from typing import Optional, List
from langchain.llms.base import LLM

from models.loader import LoaderCheckPoint
from models.base import (BaseAnswer,
                         AnswerResult,
                         AnswerResultStream,
                         AnswerResultQueueSentinelTokenListenerQueue)


class FastChatLLM(BaseAnswer, LLM, ABC):
    max_token: int = 10000
    temperature: float = 0.01
    top_p = 0.9
    checkPoint: LoaderCheckPoint = None
    # history = []
    history_len: int = 10

    def __init__(self, checkPoint: LoaderCheckPoint = None):
        super().__init__()
        self.checkPoint = checkPoint

    @property
    def _llm_type(self) -> str:
        return "FastChat"

    @property
    def _check_point(self) -> LoaderCheckPoint:
        return self.checkPoint

    @property
    def _history_len(self) -> int:
        return self.history_len

    def set_history_len(self, history_len: int = 10) -> None:
        self.history_len = history_len

    def _call(self, prompt: str, stop: Optional[List[str]] = None) -> str:
        pass

    def _generate_answer(self, prompt: str,
                         history: List[List[str]] = [],
                         streaming: bool = False,
                         generate_with_callback: AnswerResultStream = None) -> None:

        response = "fastchat 响应结果"
        history += [[prompt, response]]
        answer_result = AnswerResult()
        answer_result.history = history
        answer_result.llm_output = {"answer": response}

        generate_with_callback(answer_result)