StreamSW

Official PyTorch implementation for paper: Streaming Sliced Optimal Transport

Details of the model architecture and experimental results can be found in our papers.

@article{nguyen2025stream,
  title={Streaming Sliced Optimal Transport},
  author={Khai Nguyen},
  year={2025},
  journal={arXiv preprint arXiv:2505.06835},
}

Please CITE our paper whenever this repository is used to help produce published results or incorporated into other software.

This implementation is made by Khai Nguyen.

Requirements

To install the required python packages, run

pip install -r requirements.txt

What is included?

• Gaussian Comparison
• Mixture of Gaussians Comparison
• Gradient Flow
• Abnormality Detection

Gaussian (Mixture of Gaussians) Comparison

cd Gaussian (MixtureGaussian)
python vary_k.py
python vary_n.py
python plot_figure.py

Gradient Flow

cd GradientFlow
python main.py

Abnormality Detection

cd AbnormalityDetection
python main.py

Acknowledgment

The implementation of the KKL sketch is taken from streaming-quantiles.