app.py

import streamlit as st 
from langchain_community.llms import HuggingFaceTextGenInference
import os
from langchain.callbacks.streaming_stdout import StreamingStdOutCallbackHandler
from langchain.schema import StrOutputParser
# from datetime import datetime
from datetime import datetime, timezone, timedelta

from custom_llm import CustomLLM, custom_chain_with_history

from typing import Optional

from langchain.prompts import ChatPromptTemplate, MessagesPlaceholder
from langchain_core.chat_history import BaseChatMessageHistory
from langchain.memory import ConversationBufferMemory#, PostgresChatMessageHistory

import psycopg2
import urllib.parse as up

# os.environ['LANGCHAIN_TRACING_V2'] = "true"


API_TOKEN = os.getenv('HF_INFER_API')

@st.cache_resource
def get_llm_chain():
    return custom_chain_with_history(
        llm=CustomLLM(repo_id="AdaptLLM/medicine-LLM", model_type='text-generation', api_token=API_TOKEN, stop=["\n<|","<|"], temperature=0.001), 
        # llm=CustomLLM(repo_id="google/gemma-7b", model_type='text-generation', api_token=API_TOKEN, stop=["\n<|","<|"], temperature=0.001), 
        # memory=st.session_state.memory.chat_memory,
        memory=st.session_state.memory
    )


if 'memory' not in st.session_state:
    st.session_state['memory'] = ConversationBufferMemory(return_messages=True)
    
    # st.session_state.memory = PostgresChatMessageHistory(connection_string=POSTGRE_URL, session_id=str(datetime.timestamp(datetime.now())))

    # st.session_state.memory = get_memory()
    st.session_state.memory.chat_memory.add_ai_message("Hello, I'm AI medical consultant. How can I help you today?")
    # st.session_state.memory.add_ai_message("Hello, My name is Jonathan Jordan. You can call me Jojo. How can I help you today?")

if 'chain' not in st.session_state:
    # st.session_state['chain'] = custom_chain_with_history(
    #     llm=CustomLLM(repo_id="mistralai/Mixtral-8x7B-Instruct-v0.1", model_type='text-generation', api_token=API_TOKEN, stop=["\n<|","<|"], temperature=0.001), 
    #     memory=st.session_state.memory.chat_memory,
    #     # memory=st.session_state.memory
    # )

    st.session_state['chain'] = get_llm_chain()



st.title("AI Medical Consultation")
st.subheader("")

# Initialize chat history
if "messages" not in st.session_state:
    st.session_state.messages = [{"role":"assistant", "content":"Hello, I'm AI medical consultant. How can I help you today?"}]

# Display chat messages from history on app rerun
for message in st.session_state.messages:
    with st.chat_message(message["role"]):
        st.markdown(message["content"])

# React to user input
if prompt := st.chat_input("Ask me anything.."):
    # Display user message in chat message container
    st.chat_message("User").markdown(prompt)
    # Add user message to chat history
    st.session_state.messages.append({"role": "User", "content": prompt})
    
    response = st.session_state.chain.invoke({"question":prompt, "memory":st.session_state.memory}).split("\n<|")[0]

    # Display assistant response in chat message container
    with st.chat_message("assistant"):
        st.markdown(response)
        
    # st.session_state.memory.add_user_message(prompt)
    # st.session_state.memory.add_ai_message(response)
    st.session_state.memory.save_context({"question":prompt}, {"output":response})
    st.session_state.memory.chat_memory.messages = st.session_state.memory.chat_memory.messages[-15:]
    # Add assistant response to chat history
    st.session_state.messages.append({"role": "assistant", "content": response})