UZH ETH UZH|ETH
  Zurich

Skip to content. | Skip to navigation

Sections
Personal tools
You are here: Home Projects EvoSpike EvoSpikeSimulator

EvoSpikeSimulator

by Nikola Kasabov — last modified Mar 07, 2012 02:51 AM
Along with developing theoretical models and their pilot applications in engineering and medicine, the project is developing a software simulator called EvoSpikeSim.

EvoSpikeSim is a collection of modules and functions written in Python language and using functions from the Brian library.

EvoSpikeSim includes modules and functions as folloows:

- Modules for converting continuous-value input data into spike trains;

- Evolving spiking neural network (eSNN) models for spatio-temporal pattern recognition, including: rank-order coding eSNN; spike-time coding eSNN; eSNN with dynamic SDSP synapses; spike pattern association neuron (SPAN) and neural network models; models for classification; probabilistic eSNN models (peSNN); gene-regulatory network models for peSNN parameter optimisation; reservoir computing models; other.

- Functions for knowledge extraction from trained eSNN. 

- Functions for presenting results and for visualisation of learning processes in the peSNN.

- Functions for connecting software modules with neuromorphic hardware realisations.

A preliminary version of a visualisation tool has been developed in KEDRI by Dr Stefan Schliebs and Johannes Bopp. The tool visualises the activity in time of  120,000 LIF neurons, connected with recurrent connections in a reservoir structure, when spatio-temporal data is entered. A demo example (see the clip in the Methods and Results section shows how and when the neurons fire when boxing movement data is entered. The reservoir has 30x40x100 neurons to capture moving images of 30x40 pixels over 100 time points.              

 

The KEDRI_EvoSpike simulator code in Python along with the description for us can be downloaded from:  

http://dl.dropbox.com/u/38933969/kedri_evospike.zip