NeuralTrace is an open-source Autograd engine designed to empower deep learning enthusiasts with a powerful and user-friendly tool for training deep neural networks. It is a lightweight simulation of PyTorch.
- Automatic Differentiation: NeuralTrace performs automatic differentiation, computing gradients for various functions with respect to their inputs.
- Extensible: You can easily extend NeuralTrace by adding custom functions, optimizers, and other components.
- User-Friendly: Designed to be user-friendly, NeuralTrace is suitable for both beginners and advanced researchers.
Check out the driver.py file in the repository for more detailed examples of how to use NeuralTrace for various deep-learning tasks.
Contributions to NeuralTrace are welcome! Whether you find bugs, want to add new features, or improve documentation, please feel free to open issues and pull requests on the GitHub repository.
NeuralTrace is inspired by micrograd and based upon PyTorch.
Happy deep learning with NeuralTrace! 🚀