emeansR: Mini-Batch k-Means Clustering with Exponential Moving Averages

An R implementation of e-means (Bessa et al., 2026, doi:10.48550/arXiv.2611.09501 ), a mini-batch replacement for Lloyd's k-means that updates centroids with an exponential moving average and sets batch size and step size from fitted scaling laws. Reaches a lower loss than 25 Lloyd iterations in a few epochs. The compute core is written in Rust.

See also

Useful links:

• https://toledoem.github.io/emeansR/

• https://github.com/ToledoEM/emeansR

• Report bugs at https://github.com/ToledoEM/emeansR/issues

Author

Maintainer: Enrique Toledo enriquetoledo@gmail.com

Authors:

• Enrique Toledo enriquetoledo@gmail.com

Other contributors:

• Authors of the bundled Rust crates (see inst/AUTHORS) [contributor, copyright holder]