aathmant/htgsom

Transience GSOM implemented in Python. Supplementary code for the publication "Hierarchical Two-Stream Growing Self-Organizing Maps with Transience for Human Activity Recognition".

Transience Growing Self-Organizing Map

Inspired by the topography preserving Self-Organizing Map (SOM) proposed by Kohonen, we introduce our novel adaptation - Growing Self-Organizing Maps with Transience.

Required Modules

• Python 3
• scipy
• tqdm

Getting Started

The supplementary code repository provides the generalized python implementation of the algorithm, that can be used for dual-stream unsupervised classification. The algorithm was explained using the example Zoo dataset in the example folder.