app.py

# You can find this code for Chainlit python streaming here (https://docs.chainlit.io/concepts/streaming/python)

# OpenAI Chat completion
import os
import sys
from openai import AsyncOpenAI  # importing openai for API usage
import chainlit as cl  # importing chainlit for our app
from chainlit.prompt import Prompt, PromptMessage  # importing prompt tools
# from chainlit.playground.providers import ChatOpenAI  # importing ChatOpenAI tools
from aimakerspace.openai_utils.chatmodel import ChatOpenAI

from dotenv import load_dotenv

load_dotenv()

# Add path to the root of the repo to the system path
sys.path.append(".")

from rag import _build_vector_db, RetrievalAugmentedQAPipeline

# ChatOpenAI Templates
system_template = """You are a helpful assistant who always speaks in a pleasant tone!
"""

user_template = """{input}
Think through your response step by step.
"""


@cl.on_chat_start  # marks a function that will be executed at the start of a user session
async def start_chat():
    settings = {
        "model": "gpt-3.5-turbo",
        "temperature": 0,
        "max_tokens": 500,
        "top_p": 1,
        "frequency_penalty": 0,
        "presence_penalty": 0,
    }

    cl.user_session.set("settings", settings)


@cl.on_message  # marks a function that should be run each time the chatbot receives a message from a user
async def main(message: cl.Message):
    settings = cl.user_session.get("settings")

    # client = AsyncOpenAI()
    chat_openai = ChatOpenAI()

    print(message.content)

    vector_db = _build_vector_db()
    pipeline = RetrievalAugmentedQAPipeline(
        llm=chat_openai,
        vector_db_retriever=vector_db
    )
    response = pipeline.run_pipeline(message.content)
    
    msg = cl.Message(content="")

    # Update the prompt object with the completion
    msg.prompt = response

    # Send and close the message stream
    await msg.send()