Document Q&A Assistant

A Streamlit-based application that uses RAG (Retrieval-Augmented Generation) to answer questions about your PDF documents. The application processes your documents, creates embeddings, and uses them to provide accurate answers with source citations.

Features

• Upload and process PDF documents
• Interactive chat interface
• Source citations with viewable context
• Hybrid search for better document retrieval
• Error handling and progress tracking

Prerequisites

• Python 3.8 or higher
• pip (Python package installer)
• Virtual environment (recommended)

Installation

1. Clone the repository:

   git clone https://github.com/jmiano/pdf-chat.git
   cd pdf-chat

2. Create and activate a virtual environment:

   python -m venv venv
   source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install dependencies:

   pip install -r requirements.txt

Usage

1. Place your PDF documents in the data/ directory

2. Build the document index:

   python src/build_index.py

3. Run the Streamlit app:

   streamlit run src/app.py

4. Open your browser and navigate to the URL shown in the terminal (usually http://localhost:8501)

Project Structure

pdf-chat/
├── data/              # Directory for PDF documents
├── src/               # Source code
│   ├── app.py         # Main Streamlit application
│   └── build_index.py # Document indexing script
├── tests/             # Unit tests
├── .env               # Environment variables (create this file)
├── .gitignore         # Git ignore rules
└── requirements.txt   # Python dependencies

Configuration

Create a .env file in the root directory with your configuration:

# Example configuration
OPENAI_API_KEY=your_api_key_here

Development

Running Tests

python -m pytest tests/ -v

Code Style

The code follows PEP 8 guidelines and includes comprehensive docstrings.

License

This project is licensed under the MIT License - see the LICENSE file for details.

Contributing

1. Fork the repository
2. Create a feature branch
3. Commit your changes
4. Push to the branch
5. Create a Pull Request