VisualNnet — Hard Competitive Learning & Dot Product SOM, animated phase by phase.

Master's thesis, Czech Technical University in Prague, Faculty of Electrical Engineering, 2001 — Petr Hummel.

Core ported from the original Java applet to HTML5 Canvas in 2026 with an AI coding assistant; algorithm logic follows the original classes.

Example

Input data

Selection of input samples

Animation speed:

Fast mode (final state of each iteration only)

Seed (applied on Restart)

input samples HCL neurons Dot Product SOM neurons (weight vectors) selected sample / winning neuron circle of radius 1 (normalised vectors)