This is a command-line implementation in Go of the classic Growing Neural Gas Network algorithm described in A Growing Neural Gas Network Learns Topologies by Bernd Fritzke.
If the signal is in 2-D, the resulting topology can be drawn with the provided script plot_graph.py, and if the signal is in 3D the result can be drawn with plot_graph3D.py.
Requirements for plot_graph :
- matplotlib
- networkx
Requirements for plot_graph3D :
- networkx
- numpy
- vtk
- mayavi
- Enthought Tool Suite : mayavi, traits, traitsui, envisage, apptools
- configobj
plot_graph[3D].py [topology json file]
If the json file is not provided, the script tries to read the topology from the standard input. It is therefore possible to pipe the output of gogng directly in plot_graph.py.