app.py

import streamlit as st
from langchain_core.runnables.history import RunnableWithMessageHistory
from langchain_core.chat_history import BaseChatMessageHistory
from langchain_community.chat_message_histories import ChatMessageHistory
from langchain_google_genai import GoogleGenerativeAIEmbeddings, ChatGoogleGenerativeAI
from langchain_community.document_loaders import PyMuPDFLoader
from langchain_text_splitters import RecursiveCharacterTextSplitter
from langchain_chroma import Chroma
from langchain.chains import create_history_aware_retriever, create_retrieval_chain
from langchain.chains.combine_documents import create_stuff_documents_chain
from langchain.prompts.chat import MessagesPlaceholder, ChatPromptTemplate
import os

#  Google API key

os.environ["GOOGLE_API_KEY"]=os.getenv("TOKEN")


# Initialize the Gemini embeddings model and chat model
gemini_embeddings = GoogleGenerativeAIEmbeddings(model="models/embedding-001")
model = ChatGoogleGenerativeAI(model="gemini-1.0-pro", convert_system_message_to_human=True)

# Load and process PDF document
pdf_loader = PyMuPDFLoader(file_path="./Debyez detialed proposal 2.pdf")  # Ensure this file is in your Space
doc = pdf_loader.load()
text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200)
splits = text_splitter.split_documents(doc)

# Create a vector store and retriever



# Use Chroma in-memory mode for temporary storage
vectorstore = Chroma.from_documents(documents=splits, embedding=gemini_embeddings, persist_directory=None)
retriever = vectorstore.as_retriever()






from langchain.prompts.chat import ChatPromptTemplate, MessagesPlaceholder

# Define system prompt, ensuring that `context` is specified as an input
system_prompt = (
    "You are an assistant for question-answering tasks. "
    "Use the following pieces of retrieved context to answer the question. "
    "If you don't know the answer, say that you don't know. "
    "Use three sentences maximum and keep the answer concise."
    "\n\n"
    "{context}"
)

qa_prompt = ChatPromptTemplate.from_messages(
    [
        ("system", system_prompt),
        MessagesPlaceholder("chat_history"),
        ("human", "{input}"),
    ]
)

# Ensure the question_answer_chain uses 'context' as an expected input variable
question_answer_chain = create_stuff_documents_chain(model, qa_prompt, document_variable_name="context")

retriever_prompt = (
    "Given a chat history and the latest user question which might reference context in the chat history,"
    "formulate a standalone question which can be understood without the chat history."
)

contextualize_q_prompt = ChatPromptTemplate.from_messages(
    [
        ("system", retriever_prompt),
        MessagesPlaceholder(variable_name="chat_history"),
        ("human", "{input}"),
    ]
)

history_aware_retriever = create_history_aware_retriever(model, retriever, contextualize_q_prompt)
qa_prompt = ChatPromptTemplate.from_messages(
    [
        ("system", system_prompt),
        MessagesPlaceholder("chat_history"),
        ("human", "{input}"),
    ]
)

question_answer_chain = create_stuff_documents_chain(model, qa_prompt)
rag_chain = create_retrieval_chain(history_aware_retriever, question_answer_chain)

store = {}
def get_session_history(session_id: str) -> BaseChatMessageHistory:
    if session_id not in store:
        store[session_id] = ChatMessageHistory()
    return store[session_id]

conversational_rag_chain = RunnableWithMessageHistory(
    rag_chain,
    get_session_history,
    input_messages_key="input",
    history_messages_key="chat_history",
    output_messages_key="answer",
)

# Streamlit UI
st.title("Conversational RAG Assistant")
st.write("Ask questions based on the context in the uploaded PDF document.")

# Session ID for chat history
session_id = "user_session"

# Display chat history
if "history" not in st.session_state:
    st.session_state["history"] = []

# Get user input
user_input = st.text_input("Enter your question:", "")

if st.button("Submit"):
    if user_input:
        # Update chat history
        st.session_state["history"].append({"role": "user", "content": user_input})
        
        # Query the RAG chain
        result = conversational_rag_chain.invoke(
            {"input": user_input},
            config={"configurable": {"session_id": session_id}}
        )["answer"]
        
        # Append model response to history
        st.session_state["history"].append({"role": "assistant", "content": result})
        
        # Clear the input box after submitting
        user_input = ""

# Display conversation history
for message in st.session_state["history"]:
    if message["role"] == "user":
        st.markdown(f"**You:** {message['content']}")
    else:
        st.markdown(f"**Assistant:** {message['content']}")