app.py

import os
import streamlit as st
from PyPDF2 import PdfReader
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain.embeddings import HuggingFaceEmbeddings
from langchain.vectorstores import FAISS
from groq import Groq

# Set up Groq client
client = Groq(
    api_key="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"
)

# Streamlit app
st.title("RAG-based PDF QA Application")

# Step 1: Upload PDF document
uploaded_file = st.file_uploader("Upload a PDF document", type="pdf")

if uploaded_file:
    # Step 2: Extract text from PDF
    try:
        pdf_reader = PdfReader(uploaded_file)
        text = "\n".join(
            page.extract_text() for page in pdf_reader.pages if page.extract_text()
        )
    except Exception as e:
        st.error(f"Failed to read PDF: {e}")
        text = ""

    if text:
        # Step 3: Split text into chunks
        text_splitter = RecursiveCharacterTextSplitter(
            chunk_size=1000, chunk_overlap=200
        )
        chunks = text_splitter.split_text(text)

        # Step 4: Generate embeddings
        st.text("Generating embeddings...")
        try:
            embeddings = HuggingFaceEmbeddings(model_name="sentence-transformers/all-MiniLM-L6-v2")
            vector_db = FAISS.from_texts(chunks, embeddings)
            st.success("Embeddings generated and stored in vector database.")
        except Exception as e:
            st.error(f"Error generating embeddings: {e}")

        # Step 5: User interaction
        query = st.text_input("Ask a question based on the uploaded document:")
        if query:
            try:
                # Retrieve relevant chunks from vector DB
                docs = vector_db.similarity_search(query, k=3)
                context = "\n".join(doc.page_content for doc in docs)

                # Use Groq API for response generation
                chat_completion = client.chat.completions.create(
                    messages=[
                        {"role": "system", "content": "You are a helpful assistant."},
                        {"role": "user", "content": query},
                        {"role": "assistant", "content": context},
                    ],
                    model="llama3-8b-8192",
                    stream=False,
                )

                answer = chat_completion.choices[0].message.content
                st.text_area("Answer:", value=answer, height=200)
            except Exception as e:
                st.error(f"Error processing query: {e}")

# Footer
st.caption("Powered by Open Source Models and Groq API.")