PIT (Physics-Informed Toolbox) is a project aimed at facilitating physics-informed or constrained learning for dynamics and control applications. The toolbox provides a set of tools and resources to assist researchers and practitioners in incorporating domain knowledge and physical laws into machine learning models.
- GPU accelerated Physics-informed learning algorithms
- Batched evaluation of dynamics
- PyTorch support
To get started with PIT, follow these steps:
- Clone the repository:
git clone https://github.com/nandantumu/PIT.git - Install the required dependencies:
pip install -r requirements.txt - Explore the examples and documentation provided in the
bindirectory. - Customize and adapt the toolbox to your specific dynamics or control problem.
Contributions to PIT are welcome!