2 papers
astro-ph.IM2026
SpectralUnmix: A Torch-Based Regularized Non-negative Matrix Factorization
Rafael S. de Souza, Paula Coelho, Niranjana P +2
We present SpectralUnmix, an R package for regularized non-negative matrix factorization (NMF), implemented in torch with optional GPU acceleration. The package estimates low-rank…
astro-ph.GA2025
Capivara: A Spectral-based Segmentation Method for IFU Data Cubes
Rafael S. de Souza, Luis G. Dahmer-Hahn, Shiyin Shen +7
We present capivara, a fast and scalable spectral-based segmentation package designed to study astrophysical properties within distinct structural components of galaxies. This spec…