Somoclu: An Efficient Parallel Library for Self-Organizing Maps

Abstract

Somoclu is a massively parallel tool for training self-organizing maps on large data sets written in C++. It builds on OpenMP for multicore execution, and on MPI for distributing the workload across the nodes in a cluster. It is also able to boost training by using CUDA if graphics processing units are available. A sparse kernel is included, which is useful for high-dimensional but sparse data, such as the vector spaces common in text mining workflows. Python, R and MATLAB interfaces facilitate interactive use. Apart from fast execution, memory use is highly optimized, enabling training large emergent maps even on a single computer.

26 pages, 9 figures. The code is available at https://peterwittek.github.io/somoclu/

Cited by in corpus (6)

• High-redshift quasars at $z\ge 3$: radio variability and MPS/GPS candidates
• GEMA: An open-source Python library for self-organizing-maps
• KiDS-Legacy: Angular galaxy clustering from deep surveys with complex selection effects
• Enhancing Photometric Redshift Catalogs Through Color-Space Analysis: Application to KiDS-Bright Galaxies
• Cluster analysis of the Roma-BZCAT blazars
• Optimising the shape of photometric redshift distributions with clustering cross-correlations