Modifications of the Self-Organizing Map Program Package to enable large input data vectors

This document outlines changes in the Self-Organizing Map Program Package (SOM_PAK; Kohonen et al. 1996) designed to allow for large input data vectors. This change facilitates the production of time-evolving SOMs (Gu and Gervais, 2022) where input data vectors are three dimensional fields with both...

Full description

Bibliographic Details
Main Authors: Gu, Qinxue, Gervais, Melissa
Format: Article in Journal/Newspaper
Language:unknown
Published: Zenodo 2022
Subjects:
Online Access:https://dx.doi.org/10.5281/zenodo.6397644
https://zenodo.org/record/6397644
Description
Summary:This document outlines changes in the Self-Organizing Map Program Package (SOM_PAK; Kohonen et al. 1996) designed to allow for large input data vectors. This change facilitates the production of time-evolving SOMs (Gu and Gervais, 2022) where input data vectors are three dimensional fields with both space and time dimensions. The original SOM_PAK is available at the following link: http://www.cis.hut.fi/research/som_lvq_pak.shtml. This document is associated a journal article (Gu and Gervais, 2022). Please cite both Kohonen et al. (1996) and Gu and Gervais (2022) when using this method. References: Kohonen, T. K., Hynninen, J., Kangas, J., & Laaksonen, J. (1996). SOM PAK: The Self-Organizing Map Program Package. Report A31, Helsinki University of Technology, Laboratory of Computer and Information Science, February. http://www.cis.hut.fi/nnrc/nnrc-programs.html Gu, Q., & Gervais, M. (2022). Diagnosing two-way coupling in decadal North Atlantic SST variability using time-evolving self-organizing maps. Geophysical Research Letters