import streamlit as st 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 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-chat", model_type='text-generation', api_token=API_TOKEN, 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})