Welcome to Cheetah’s documentation!

Cheetah is a particle tracking accelerator we built specifically to speed up the training of reinforcement learning models.

GitHub repository: https://github.com/desy-ml/cheetah

Paper: https://arxiv.org/abs/2401.05815


Simply install Cheetah from PyPI by running the following command.

pip install cheetah-accelerator


We provide some examples to demonstrate some features of Cheetah and show how to use them. They provide a good entry point to using Cheetah, but they do not represent its full functionality. To move beyond the examples, please refer to the in-depth documentation. If you feel like other examples should be added, feel free to open an issue on GitHub.


For more advanced usage, please refer to the in-depth documentation.

Cite Cheetah

If you use Cheetah, please cite the following two papers:

    title        = {Bridging the gap between machine learning and particle accelerator physics with high-speed, differentiable simulations},
    author       = {Kaiser, Jan and Xu, Chenran and Eichler, Annika and Santamaria Garcia, Andrea},
    year         = 2024,
    month        = {May},
    journal      = {Phys. Rev. Accel. Beams},
    publisher    = {American Physical Society},
    volume       = 27,
    pages        = {054601},
    doi          = {10.1103/PhysRevAccelBeams.27.054601},
    url          = {https://link.aps.org/doi/10.1103/PhysRevAccelBeams.27.054601},
    issue        = 5,
    numpages     = 17
    title        = {Accelerating Linear Beam Dynamics Simulations for Machine Learning Applications},
    author       = {Stein, Oliver and Kaiser, Jan and Eichler, Annika},
    year         = 2022,
    booktitle    = {Proceedings of the 13th International Particle Accelerator Conference}

For Developers

Activate your virtual environment. (Optional)

Install the cheetah package as editable

pip install -e .

We suggest installing pre-commit hooks to automatically conform with the code formatting in commits:

pip install pre-commit
pre-commit install

Indices and tables