Micrograd

This project is a neural network implementation inspired by Andrej Karpathy's "Neural Networks: Zero to Hero" video series. The main goal is to implement deep learning concepts from scratch in a clear and understandable way.

Project Features

• Autograd engine for automatic differentiation
• Basic neural network components
• Visualization tools
• Computational graph creation
• Backpropagation implementation

Quick Start

# Clone the repository
git clone https://github.com/aynursusuz/micrograd.git
cd micrograd

# Install dependencies
pip install -r requirements.txt

Requirements

• Python 3.8+
• NumPy
• Matplotlib
• Graphviz (for visualization)
• Jupyter Notebook

Project Structure

micrograd/
├── notebooks/
│   └── 01_micrograd.ipynb  # Core examples and explanations
├── code/
│   ├── engine.py          # Autograd engine
│   └── utils.py           # Helper functions
└── requirements.txt

Learning Resources

• Neural Networks: Zero to Hero Video Series
• Automatic Differentiation Concepts
• Backpropagation Algorithm

Contributing

We welcome your suggestions and pull requests to improve the project! You can contribute by adding new features, fixing bugs, or improving documentation.

License

MIT License 2024

This project is open source and available under the MIT License.