app.py

import os
import streamlit as st
import pdfplumber
from sentence_transformers import SentenceTransformer
import faiss
import numpy as np
from groq import Groq

# Set background image and customize colors
background_image_url = "https://cdn.pixabay.com/photo/2016/06/02/02/33/triangles-1430105_1280.png"
st.markdown(
    f"""
    <style>
    .stApp {{
        background-image: url("{background_image_url}");
        background-size: cover;
        background-position: center center;
        background-repeat: no-repeat;
    }}
    
    /* Ensure title is black */
    h1 {{
        color: black !important;  /* Force title color to black */
    }}
    
    /* Set footer text color to white */
    h2, h3, h4, h5, h6, p {{
        color: white;  /* Set all text color to white */
    }}
    
    /* Set footer styling */
    .footer {{
        position: fixed;
        bottom: 0;
        left: 0;
        right: 0;
        background-color: rgba(0, 0, 0, 0.6);
        color: white;
        text-align: center;
        padding: 10px 0;
        font-size: 14px;
    }}
    
    /* Set processing button color to green */
    .stButton button {{
        background-color: green;
        color: white;
    }}
    /* Set query input block background color to white */
    .stTextInput input {{
        background-color: white;
        color: black;
        border-radius: 5px;
        padding: 10px;
    }}
    
    /* Set all output text (retrieved chunks and responses) to white */
    .stMarkdown, .stTextInput, .stText, .stCode, .stJson, .stFileUploader, .stError, .stSuccess {{
        color: white !important;
    }}
    </style>
    """, 
    unsafe_allow_html=True
)

# Use your Groq API key from Hugging Face Secrets
HUGGINGFACE_KEY = os.getenv("HUGGINGFACE_KEY")

if not HUGGINGFACE_KEY:
    st.error("Groq API key not found. Please set it in Hugging Face Secrets.")

# Initialize Groq client with the correct API key
groq_client = Groq(api_key=HUGGINGFACE_KEY)

# Load the SentenceTransformer model for embedding generation
embedder = SentenceTransformer('sentence-transformers/all-MiniLM-L6-v2')

# Define file path and vector store folder
file_path = "The Rise of Agentic AI.pdf"  # File directly in the root directory of the app
VECTORSTORE_FOLDER = "vectorstore"  # Folder where the FAISS index will be stored

# Ensure the vector store folder exists
if not os.path.exists(VECTORSTORE_FOLDER):
    os.makedirs(VECTORSTORE_FOLDER)

# Define the vector store path
vectorstore_path = os.path.join(VECTORSTORE_FOLDER, "index.faiss")  # Correct path to the index file

# Load or create FAISS index
if os.path.exists(vectorstore_path):
    # If the index file exists, read it
    try:
        index = faiss.read_index(vectorstore_path)
    except Exception as e:
        st.error(f"Error reading the FAISS index: {e}")
        index = faiss.IndexFlatL2(embedder.get_sentence_embedding_dimension())
else:
    # If the index file doesn't exist, create a new one
    index = faiss.IndexFlatL2(embedder.get_sentence_embedding_dimension())

# Variable to hold chunks globally
chunks = []

# Function to load text from PDF
def load_pdf_text(file_path):
    """Extract text from the given PDF file."""
    text = ""
    with pdfplumber.open(file_path) as pdf:
        for page in pdf.pages:
            text += page.extract_text()
    return text

# Function to chunk text into smaller pieces
def chunk_text(text, chunk_size=500, overlap=100):
    """Chunk the text into overlapping chunks."""
    chunks = []
    for i in range(0, len(text), chunk_size - overlap):
        chunks.append(text[i:i + chunk_size])
    return chunks

# Process the document and update vector store
def process_and_store_document(file_path):
    """Process the PDF document, chunk text, generate embeddings, and store them in FAISS."""
    global chunks  # Make chunks global to access in the query part
    
    st.info("Processing PDF document...")
    
    # Extract text from the PDF file
    text = load_pdf_text(file_path)
    
    # Chunk the text into smaller pieces
    chunks = chunk_text(text)
    
    # Generate embeddings for each chunk
    embeddings = embedder.encode(chunks, show_progress_bar=True)
    
    # Add the embeddings to the FAISS index
    index.add(np.array(embeddings))
    
    # Save the updated FAISS index
    try:
        faiss.write_index(index, vectorstore_path)
        st.success("Document processed and vector store updated!")
    except Exception as e:
        st.error(f"Error saving the FAISS index: {e}")

# User interface for Streamlit
st.title("The Rise of Agentic AI RAG Application")

# Button to trigger document processing
if st.button("Process PDF"):
    process_and_store_document(file_path)

# Query input for the user
user_query = st.text_input("Enter your query:", key="query_input")

if user_query:
    # Check if there are any chunks in the index
    if not chunks:
        st.error("Please process the document first by clicking 'Process PDF'.")
    else:
        # Generate embedding for the user query
        query_embedding = embedder.encode([user_query])
        
        # Perform the search on the FAISS index
        distances, indices = index.search(np.array(query_embedding), k=5)
        
        # Check if the indices returned are valid
        if indices.size == 0 or np.any(indices[0] == -1):
            st.error("No relevant results found in the index.")
        else:
            # Ensure indices are within the bounds of the chunks list
            valid_indices = [idx for idx in indices[0] if idx < len(chunks)]
            
            if not valid_indices:
                st.error("No valid indices found for the retrieved chunks.")
            else:
                # Retrieve the most relevant chunks based on the valid indices
                retrieved_chunks = [chunks[idx] for idx in valid_indices]
                
                # Combine the retrieved chunks with the query and generate a response using Groq
                combined_input = " ".join(retrieved_chunks) + user_query
                
                # Generate a response with Groq
                try:
                    chat_completion = groq_client.chat.completions.create(
                        messages=[{
                            "role": "user",
                            "content": combined_input,
                        }],
                        model="llama3-8b-8192",  # Specify the model you want to use
                    )

                    # Display only the generated response
                    st.subheader("Generated Response")
                    st.write(chat_completion.choices[0].message.content)
                except Exception as e:
                    st.error(f"Error generating response: {e}")

# Footer
st.markdown("<div class='footer'>Created by Shamil Shahbaz</div>", unsafe_allow_html=True)