简单的chatGPT 网页界面

        大语言模型的应用大多数采用python 语言实现，为了与其它应用结合，最简单的方法就是采样网页 RESTful API 服务。目前流行的Python 的Web 服务是FastAPI。本文记录了一个简单LLM 对话FastAPI 的网站服务的实验过程。

界面

安装

安装如下两个主要模块

pip install fastapi 
pip install uvicorn

文件目录结构

FastAPI 具有固定的目录结构，Templates 中包含了index.hml

static 文件夹

,static 中包含了js,css,images等文件夹。

主程序（main.py)

import asyncio
import nest_asyncio
from langchain_core.prompts import PromptTemplate
from langchain.chains import LLMChain
from langchain.chat_models import ErnieBotChat
from pydantic import BaseModel
nest_asyncio.apply()
llm= ErnieBotChat(model_name='ERNIE-Bot', #ERNIE-Bot
                    ernie_client_id='xxxxxxxx',
                    ernie_client_secret='xxxxxxxxxxx',
                    temperature=0.75,
                    )
template = """You are a nice chatbot having a conversation with a human.
New human question: {question}
Response:"""
prompt = PromptTemplate.from_template(template)
# Notice that we need to align the `memory_key`
conversation = LLMChain(
    llm=llm,
    prompt=prompt,
    verbose=True,
)
class Prompt(BaseModel):
    Method: str
    Message:str  
    
    
app = FastAPI()
app.mount("/static", StaticFiles(directory="static"), name="static")
templates = Jinja2Templates(directory="templates")
@app.get("/")
async def root(request: Request):
   # return {"message": "Hello, World!"}
 
   return templates.TemplateResponse("index.html",{
            "request": request
        })
@app.post("/generate/")
def generate(prompt:Prompt):
    print(prompt)
    AIresponse=conversation.predict(question=prompt.Message)
    response=prompt
    response.Message=AIresponse
    print(response)
    return {"response": response}
async def run_server():
    uvicorn.run(app, host="localhost", port=8000)

if __name__ == "__main__":
    loop = asyncio.get_event_loop()
    loop.run_until_c