🤖 AlgoAssist: Data Structures & Algorithms RAG Chatbot

AlgoAssist is an intelligent Retrieval-Augmented Generation (RAG) chatbot designed to help students and developers learn and explore Data Structures and Algorithms (DSA). It leverages the power of LangChain, ChromaDB, LLaMA3, and Streamlit to provide accurate, context-aware answers extracted from a comprehensive DSA textbook.

🚀 Features

• 📘 PDF Knowledge Base: Ingests and indexes a DSA textbook using ChromaDB and sentence-transformer embeddings
• 🔍 Retrieval-Augmented Generation: Retrieves relevant text chunks before generating responses using LLaMA3
• 💬 Natural Chat Interface: Built with Streamlit for an interactive and user-friendly experience
• ⚙️ Efficient Query Processing: Uses sentence-transformers for semantic search and LLaMA3-8B for generation
• 📊 Context-Aware Responses: Provides algorithm explanations, comparisons, and implementation details with academic precision

📋 Prerequisites

• Python 3.9+
• Groq API key (for LLaMA3 access)
• HuggingFace token (for embeddings)
• A PDF of a Data Structures and Algorithms textbook

📁 Project Structure

LLM-RAG/
├── app/
│   ├── __init__.py
│   ├── chatbot.py           # Core RAG logic
│   ├── chroma_data.py       # Builds ChromaDB from PDF
│   ├── llm.py               # Loads LLM, retriever, and prompt
├── chroma_db/               # Persistent vector store
│   ├── chroma.sqlite3
│   └── <auto-generated UUID folder>
├── data/
│   └── DSA BOOK.pdf         # Source document (DSA book)
├── interface/
│   ├── __init__.py
│   └── streamlit.py         # Streamlit frontend
├── llm-rag-env/             # Your virtual environment (not version controlled)
├── config.py                # API keys and constants
├── requirements.txt         # Python dependencies
├── run.sh                   # Shell script to start app
├── Dockerfile               # Docker support
├── .env                     # Environment variables (not version controlled)
├── .gitignore
├── .dockerignore
└── README.md                # Project documentation

🛠️ Setup Instructions

1. Clone the Repository

git clone https://github.com/harshhrawte/LLM-RAG.git
cd LLM-RAG

2. Create a Virtual Environment

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

3. Install Dependencies

pip install -r requirements.txt

4. Set Up API Keys

Create a config.py file in the root directory with the following content:

GROQ_API_KEY = "your-groq-api-key"
HF_TOKEN = "your-huggingface-token"

Visit http://localhost:8501 in your browser to start chatting with AlgoAssist!

5. Vector Database (BOOK provide for the chroma DB)

https://drive.google.com/file/d/1W65q3k0WWLJNfmbnptyBIkEQXIULi57f/view?usp=sharing

🐳 Docker Support

You can also run AlgoAssist in a Docker container:

# Build the Docker image
docker build -t algoassist .

# Run the container
docker run -p 8501:8501 algoassist

📱 Usage Examples

• "Explain the time complexity of quicksort."
• "What's the difference between BFS and DFS?"
• "How do I implement a balanced binary search tree?"
• "Compare dynamic programming and greedy algorithms."
• "Explain the Dijkstra's algorithm with an example."

🔄 How It Works

1. Text Ingestion: The DSA book is split into chunks and embedded using sentence-transformers.
2. Vector Storage: Text embeddings are stored in ChromaDB for efficient retrieval.
3. Query Processing: User questions are embedded and semantically similar chunks are retrieved.
4. Context-Enhanced Generation: LLaMA3 generates responses based on the retrieved context and the user query. 📸 Output Examples Here are some screenshots showing AlgoAssist in action:

Example 1: Explains the basic concept :

AlgoAssist explaining a complex algorithm with context from the DSA book

Example 2: Gives the relevant link to solve and practice the problem

weblink of the problem

Example 3: Solves the Complex problems

Example 4: Solves errors from the code snippet provided

Code explanation and implementation details for a specific algorithm

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

👏 Acknowledgements

LangChain for the RAG framework Groq for LLaMA3 API access HuggingFace for embedding models ChromaDB for vector storage Streamlit for the interactive interface

🔗 Contact

For questions or feedback, please open an issue on this repository or contact:

• GitHub: harshhrawte

Happy coding and learning DSA with AlgoAssist! 🚀