EEEGPT v1.0: AI Assistant for Electrical and Electronic Engineering

EEEGPT is an AI-powered assistant designed to help students and professionals in Electrical and Electronic Engineering (EEE). It leverages advanced language models and vector search to provide accurate, context-based answers from authoritative EEE textbooks.

Features

• Contextual Q&A: Answers questions using content from a trusted EEE textbook PDF.
• Beginner-Friendly: Explains concepts in simple language, with technical terms defined.
• Technical Support: Provides step-by-step reasoning, LaTeX equations, and code snippets (e.g., SPICE netlists) where relevant.
• Source Transparency: Optionally shows source document snippets for each answer.
• Interactive UI: Built with Streamlit for a conversational chat experience.

Project Structure

.
├── connect_memory_with_llm.py
├── create_memory_for_llm.py
├── eeegpt_ui.py
├── test.py
├── requirements.txt
├── README.md
├── data/
│   └── a-textbook-of-electrical-technology-volume-i-basic-electrical-engineering-b-l-theraja.pdf
└── vectorstore/
    └── db_faiss/
        ├── index.faiss
        └── index.pkl

How It Works

1. Document Loading: Loads and splits the textbook PDF into manageable text chunks.
2. Embedding Generation: Uses HuggingFace's sentence-transformers/all-MiniLM-L6-v2 to create vector embeddings.
3. Vector Store: Stores embeddings in a FAISS vector database for efficient retrieval.
4. LLM Integration: Uses Groq's Llama 3 model via API for generating answers.
5. Retrieval-Augmented Generation: Retrieves relevant context from the vector store and feeds it to the LLM for accurate, context-aware responses.

Setup Instructions

1. Clone the Repository

git clone https://github.com/Nobi004/eeegpt.git
cd eeegpt

2. Install Dependencies

pip install -r requirements.txt

3. Prepare the Data

• Place your EEE textbook PDF(s) in the data/ directory.

4. Create the Vector Store

Run the following script to process the PDF and build the FAISS vector database:

python create_memory_for_llm.py

5. Launch the Streamlit UI

streamlit run eeegpt_ui.py

Usage

• Ask any EEE-related question in the chat interface.
• The assistant will answer using only the information from the provided textbook context.
• If the context is insufficient, it will let you know.

Customization

• Add More PDFs: Place additional EEE PDFs in the data/ folder and rerun create_memory_for_llm.py.
• Prompt Engineering: Modify the prompt template in eeegpt_ui.py to adjust the assistant's behavior.
• Model Settings: Change the LLM or embedding model as needed in the code.

Troubleshooting

• Ensure all dependencies are installed.
• Make sure the FAISS database exists at vectorstore/db_faiss/.
• Verify your Groq API key is valid and set in the code.

Acknowledgements

• LangChain
• HuggingFace Transformers
• FAISS
• Streamlit
• Groq API

License

This project is for educational and research purposes only. Please ensure you have the rights to use any PDF content you provide.

---

Made by Md. Mahmudun Nobi