📄 Chat with PDFs using RAG

A Streamlit-based AI application that allows users to upload PDF documents and ask questions about their content using Retrieval-Augmented Generation (RAG). The application extracts text from PDFs, generates embeddings, stores them in ChromaDB, retrieves relevant context, and uses the Groq LLM to answer user queries.

---

🚀 Live Demo

🔗 Try the App: https://chat-with-pdfsss.streamlit.app

---

📌 Features

• 📄 Upload one or multiple PDF documents
• ✂️ Extract text using PyMuPDF
• 🧩 Split documents into semantic chunks
• 🔍 Generate embeddings for efficient retrieval
• 🗄️ Store embeddings using ChromaDB
• 🤖 Ask natural language questions about uploaded PDFs
• ⚡ Generate AI-powered answers using Groq LLM
• 💻 Interactive and user-friendly Streamlit interface

---

🛠️ Tech Stack

Technology	Purpose
Python	Programming Language
Streamlit	Web Application
ChromaDB	Vector Database
Sentence Transformers	Text Embeddings
Groq API	Large Language Model
PyMuPDF	PDF Text Extraction
python-dotenv	Environment Variables

---

📂 Project Structure

Chat-with-Pdfs/
│
├── app.py
├── requirements.txt
├── README.md
├── .gitignore
├── .env.example
│
├── src/
│   ├── embeddings.py
│   ├── ingestion.py
│   ├── qa.py
│   └── __init__.py
│
├── pdfs/
├── data/
└── logs/

---

⚙️ Installation

1. Clone the Repository

git clone https://github.com/Krishna9616/Chat-with-Pdfs.git
cd Chat-with-Pdfs

2. Create a Virtual Environment

python -m venv venv

3. Activate the Virtual Environment

Windows

venv\Scripts\activate

macOS/Linux

source venv/bin/activate

4. Install Dependencies

pip install -r requirements.txt

---

🔑 Environment Variables

Create a .env file in the project root.

Add your Groq API key:

GROQ_API_KEY=your_groq_api_key

---

▶️ Run the Application

streamlit run app.py

The application will open automatically in your browser.

---

🌐 Deployment

This project is deployed using Streamlit Community Cloud.

Steps to Deploy

1. Push the project to GitHub.
2. Connect the repository to Streamlit Community Cloud.
3. Set the Main file path as:

app.py

4. Add your API key under App Settings → Secrets:

GROQ_API_KEY="your_groq_api_key"

---

📸 Screenshots

Home Page

---

Chat Interface

---

💡 How It Works

1. User uploads one or more PDF documents.
2. The application extracts text using PyMuPDF.
3. The extracted text is divided into chunks.
4. Sentence Transformer embeddings are generated.
5. Embeddings are stored in ChromaDB.
6. When a user asks a question:
   • Similar chunks are retrieved.
   • Retrieved context is sent to the Groq LLM.
   • The AI generates an accurate answer.

---

👨‍💻 Author

Krishna

• GitHub: https://github.com/Krishna9616
• Project Repository: https://github.com/Krishna9616/Chat-with-Pdfs
• Live Demo: https://chat-with-pdfsss.streamlit.app

---