torchsom: The Reference PyTorch Library for Self-Organizing Maps

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Berthier, Louis, Shokry, Ahmed, Moreaud, Maxime, Ramelet, Guillaume, Moulines, Eric
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866909841227776000
author Berthier, Louis
Shokry, Ahmed
Moreaud, Maxime
Ramelet, Guillaume
Moulines, Eric
author_facet Berthier, Louis
Shokry, Ahmed
Moreaud, Maxime
Ramelet, Guillaume
Moulines, Eric
contents This paper introduces torchsom, an open-source Python library that provides a reference implementation of the Self-Organizing Map (SOM) in PyTorch. This package offers three main features: (i) dimensionality reduction, (ii) clustering, and (iii) friendly data visualization. It relies on a PyTorch backend, enabling (i) fast and efficient training of SOMs through GPU acceleration, and (ii) easy and scalable integrations with PyTorch ecosystem. Moreover, torchsom follows the scikit-learn API for ease of use and extensibility. The library is released under the Apache 2.0 license with 90% test coverage, and its source code and documentation are available at https://github.com/michelin/TorchSOM.
format Preprint
id arxiv_https___arxiv_org_abs_2510_11147
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle torchsom: The Reference PyTorch Library for Self-Organizing Maps
Berthier, Louis
Shokry, Ahmed
Moreaud, Maxime
Ramelet, Guillaume
Moulines, Eric
Machine Learning
68T07, 68T05, 68-04
I.2.6; I.5.1; I.5.3
This paper introduces torchsom, an open-source Python library that provides a reference implementation of the Self-Organizing Map (SOM) in PyTorch. This package offers three main features: (i) dimensionality reduction, (ii) clustering, and (iii) friendly data visualization. It relies on a PyTorch backend, enabling (i) fast and efficient training of SOMs through GPU acceleration, and (ii) easy and scalable integrations with PyTorch ecosystem. Moreover, torchsom follows the scikit-learn API for ease of use and extensibility. The library is released under the Apache 2.0 license with 90% test coverage, and its source code and documentation are available at https://github.com/michelin/TorchSOM.
title torchsom: The Reference PyTorch Library for Self-Organizing Maps
topic Machine Learning
68T07, 68T05, 68-04
I.2.6; I.5.1; I.5.3
url https://arxiv.org/abs/2510.11147