Stylor is a high-performance neural style transfer framework leveraging Intel oneDNN for efficient training and inference, enabling real-time artistic transformation with minimal computational overhead.
Stylor is distributed under the MIT License. Refer to the LICENSE file for details.
Please cite our software if you use this code or part of it in your work:
@misc{Stylor,
title={stylor},
author={Nathanael Bracy},
year={2025},
url={https://github.com/servusdei2018/stylor},
note={edge-optimized neural style transfer using Intel oneDNN}
}