RagChitChat: Retrieval Augmented Chat for PPT/PDF Notes

A powerful Retrieval Augmented Generation (RAG) chatbot designed specifically for .ppt .pdf notes. It combines local LLM capabilities with advanced document retrieval to provide accurate, context-aware answers about software engineering concepts.

demo : youtube DEMO

## 🌟 Features

• 📚 Document Processing: Intelligent handling of various file formats

  • PDF lecture notes with multi-page support
  • PowerPoint (PPTX) presentations
  • Automatic chunking with overlap for context preservation
  • Text extraction with metadata retention (source, page numbers)

• 🔍 Advanced Retrieval:

  • Hybrid search combining semantic (vector) and keyword-based (BM25) retrieval
  • ChromaDB vector database for efficient similarity search
  • Custom relevance scoring and result ranking
  • Configurable retrieval parameters (chunk size, overlap, top_k)

• 🧠 Local LLM Integration:

  • Runs entirely offline with Ollama
  • Multiple model support with live switching (mistral, llama2, etc.)
  • Optimized prompting with context window management
  • Stream responses for better user experience

• 📊 Advanced Prompt Engineering:

  • Chain-of-thought reasoning
  • Few-shot learning examples for complex queries
  • Self-reflective answer generation
  • Structured output formatting based on query type
  • Dynamic prompt selection based on question analysis

• 💻 Rich Terminal UI:

  • Beautiful interactive interface with Rich
  • Markdown rendering for formatted responses
  • Progress indicators during processing
  • Command system with help, history, and model management
  • Syntax highlighting and pretty printing

📋 Project Structure(OOP Based)

RagChitChat/
├── README.md                # Project documentation
├── requirements.txt         # Dependencies
├── setup.py                 # Package setup for imports
├── .env.example             # Environment variables template
├── .gitignore               # Git ignore rules
│
├── data/                    # Raw lecture notes (.pdf, .pptx)
├── processed/               # Processed text data
├── chroma_db/               # Vector database storage
│
├── src/                     # Source code
│   ├── __init__.py
│   ├── document_processor/  # PDF and PPTX processing
│   │   ├── __init__.py
│   │   └── processor.py     # Document processing classes
│   │
│   ├── vector_store/        # ChromaDB integration
│   │   ├── __init__.py
│   │   └── chroma_store.py  # Vector database management
│   │
│   ├── retriever/           # Haystack retrieval components
│   │   ├── __init__.py
│   │   └── haystack_retriever.py  # Document retrieval
│   │
│   ├── llm/                 # Ollama integration
│   │   ├── __init__.py
│   │   └── ollama_client.py # LLM interface
│   │
│   ├── prompts/             # Prompt engineering
│   │   ├── __init__.py
│   │   └── prompt_templates.py  # Advanced prompt templates
│   │
│   ├── interface/           # Rich terminal UI
│   │   ├── __init__.py
│   │   └── terminal_ui.py   # Terminal interface
│   │
│   └── main.py              # Entry point
│
└── config/                  # Configuration files
    ├── __init__.py
    └── settings.py          # Application settings

🛠️ Requirements

System Requirements

• Python 3.8 or higher
• 8GB RAM minimum (16GB recommended for larger models)
• Storage space for models and vector database (2GB+)
• Windows, macOS, or Linux

Required Tools

• Ollama: For running local LLMs (Download)
• Python: 3.8+ (Download)

🚀 Complete Setup Instructions

1. Install Python and Setup Virtual Environment

# Install Python (if not already installed)
# Download from https://www.python.org/downloads/

# Clone the repository (if using git)
git clone https://github.com/nxdun/RagChitChat.git
cd RagChitChat

# Create and activate virtual environment
python -m venv env

# On Windows
.\env\Scripts\activate

# On macOS/Linux
source env/bin/activate

2. Install Dependencies

# Install required packages
pip install -r requirements.txt

# (optional)Install package for imports (optional, if import issues occur)
pip install -e .

3. Install and Configure Ollama

# Download and install Ollama from https://ollama.ai/

# Pull required models (after installing)
ollama pull mistral:7b-instruct-v0.3-q4_1  # Default model(You can edit this on env)
ollama pull llama2                          # Alternative model (optional)

4. Prepare Your Data

# Create data directory (root Dir)
mkdir -p data

# Copy your lecture notes into data folder
# Supported formats: PDF, PPTX
# Example: cp ~/Downloads/CTSE_Lecture*.pdf data/

5. Configure Environment Variables (Optional : Or Use Defaults)

# Copy example environment file
cp .env.example .env

# Edit .env file with your settings
# nano .env or use any text editor

6. Run the Application

# Start the chatbot
python src/main.py

📝 Configuration Options

You can configure the application through environment variables or by editing config/settings.py:

Variable	Description	Default
RAGCHITCHAT_DATA_DIR	Directory containing lecture notes	data
RAGCHITCHAT_PROCESSED_DIR	Directory for processed documents	processed
RAGCHITCHAT_DB_DIR	Directory for vector database	chroma_db
RAGCHITCHAT_MODEL	Default Ollama model	mistral:7b-instruct-v0.3-q4_1
OLLAMA_URL	Ollama API URL	http://localhost:11434
RAGCHITCHAT_CHUNK_SIZE	Document chunk size	1000
RAGCHITCHAT_CHUNK_OVERLAP	Overlap between chunks	200
RAGCHITCHAT_TOP_K	Number of context documents to retrieve	5

🖥️ Usage Guide

Basic Commands

Once the application is running, you can interact with it through the terminal:

Command	Description
/help	Show available commands
/exit	Exit the application
/clear	Clear conversation history
/history	Show conversation history
/models	List available models
/model <name>	Switch to a different model
/info	Show system information
/about	About the application

Example Questions

Here are some examples of questions you can ask:

• "What is continuous integration?"
• "Explain the difference between DevOps and DevSecOps"
• "What are the benefits of microservices architecture?"
• "How does containerization improve software deployment?"
• "Compare agile and waterfall methodologies in software engineering"

📊 Advanced Features

Model Switching

You can switch between different LLMs during runtime:

/models                         # List available models
/model mistral:7b-instruct-v0.3-q4_1   # Switch to Mistral
/model llama2                   # Switch to Llama 2

Dynamic Prompt Selection

The system automatically selects prompting strategies based on question type:

• Factual questions: Standard RAG with direct answers
• Comparative questions: Structured comparison format
• Procedural questions: Step-by-step instructions
• Complex questions: Self-reflective generation

Hybrid Search

The retrieval system combines two search methods for better results:

1. Vector search: Semantic similarity using embeddings
2. BM25 search: Keyword-based traditional search

🔧 Troubleshooting

Common Issues and Solutions

Problem: Import errors when running python src/main.py Solution: Install the package in development mode:

pip install -e .

Problem: "Cannot connect to Ollama" error Solution: Ensure Ollama is installed and running:

# Check if Ollama is running
curl http://localhost:11434/api/tags
# If not running, start Ollama application

Problem: Model not found when switching models Solution: Pull the model using Ollama CLI:

ollama pull <model_name>

Problem: High memory usage Solution: Use a smaller model or reduce RAGCHITCHAT_TOP_K in settings

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

1. Fork the repository
2. Create your feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add some amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

📄 License

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

Acknowledgements

• Ollama for local LLM capabilities
• Haystack for the RAG framework
• ChromaDB for vector storage
• Rich for the terminal UI