app.py

import os
import streamlit as st
# Update these imports
from langchain_community.embeddings import HuggingFaceEmbeddings
from langchain.chains import RetrievalQA
from langchain_community.vectorstores import FAISS
from langchain_core.prompts import PromptTemplate
from langchain_huggingface import HuggingFaceEndpoint

from dotenv import load_dotenv, find_dotenv
load_dotenv(find_dotenv())

DB_FAISS_PATH = "vectorstore/db_faiss"

@st.cache_resource
def get_vectorstore():
    embedding_model = HuggingFaceEmbeddings(model_name='sentence-transformers/all-MiniLM-L6-v2')
    db = FAISS.load_local(DB_FAISS_PATH, embedding_model, allow_dangerous_deserialization=True)
    return db

def set_custom_prompt(custom_prompt_template):
    prompt = PromptTemplate(template=custom_prompt_template, input_variables=["context", "question"])
    return prompt

def load_llm(huggingface_repo_id, HF_TOKEN):
    llm = HuggingFaceEndpoint(
        repo_id=huggingface_repo_id,
        task="text-generation",  # Add this line
        temperature=0.5,
        model_kwargs={
            "token": HF_TOKEN,
            "max_length": 512  # Changed to integer
        }
    )
    return llm

def main():
    st.title("Forecasting discharge outcomes for critically ILL patients using machine learning!")
    
    if 'messages' not in st.session_state:
        st.session_state.messages = []
    
    for message in st.session_state.messages:
        st.chat_message(message['role']).markdown(message['content'])
    
    prompt = st.chat_input("Pass your prompt here")
    
    if prompt:
        st.chat_message('user').markdown(prompt)
        st.session_state.messages.append({'role': 'user', 'content': prompt})
        
        CUSTOM_PROMPT_TEMPLATE = """
        Use the pieces of information provided in the context to answer user's question.
        If you dont know the answer, just say that you dont know, dont try to make up an answer.
        
        Dont provide anything out of the given context
        
        Context: {context}
        Question: {question}
        
        Start the answer directly. No small talk please.
        """
        
        HUGGINGFACE_REPO_ID = "mistralai/Mistral-7B-Instruct-v0.3"
        HF_TOKEN = os.environ.get("HF_TOKEN")
        
        try:
            with st.spinner("Thinking..."):  # Add loading indicator
                vectorstore = get_vectorstore()
                if vectorstore is None:
                    st.error("Failed to load the vector store")
                    return

                qa_chain = RetrievalQA.from_chain_type(
                    llm=load_llm(huggingface_repo_id=HUGGINGFACE_REPO_ID, HF_TOKEN=HF_TOKEN),
                    chain_type="stuff",
                    retriever=vectorstore.as_retriever(search_kwargs={'k': 3}),
                    return_source_documents=True,
                    chain_type_kwargs={'prompt': set_custom_prompt(CUSTOM_PROMPT_TEMPLATE)}
                )
                
                response = qa_chain.invoke({'query': prompt})
                
                result = response["result"]
                source_documents = response["source_documents"]
                
                # Format source documents more cleanly
                source_docs_text = "\n\n**Source Documents:**\n"
                for i, doc in enumerate(source_documents, 1):
                    source_docs_text += f"{i}. Page {doc.metadata.get('page', 'N/A')}: {doc.page_content[:200]}...\n\n"
                
                result_to_show = f"{result}\n{source_docs_text}"
                
                st.chat_message('assistant').markdown(result_to_show)
                st.session_state.messages.append({'role': 'assistant', 'content': result_to_show})
                
        except Exception as e:
            st.error(f"Error: {str(e)}")
            st.error("Please check your HuggingFace token and model access permissions")

if __name__ == "__main__":
    main()