Brainlab

Brainlab is a toolkit to aid in the design, simulation, and analysis of spiking neural networks with the NCS environment. Brainlab is a set of Python libraries.

Brainlab is beta software. Use at your own risk. Brainlab can be downloaded here.

This version is released under GPL version 2, available here.

In order to make full use of Brainlab, you will probably need to install NCS, the NeoCortical Simulator. Please go here for information on obtaining NCS. It is possible to run NCS on a single computer; a Beowulf or BlueGene cluster is not required. It helps though.

Refer to my thesis for some instructions on installing and using Brainlab. Some slides from a Brainlab presentation are here.

The Brainlab online documentation might help too. Mostly it will help if you already know what you want to do.

You will need to create a .brainlabrc file in the current directory and in there at least indicate whether you want to work locally or remotely. A sample is included as dot-brainlabrc in the distribution.

Brainlab requires a number of support libraries, including:
matplotlib (for interactive displays, you will need some backend too)
numarray (Numeric might work too)
scipy (maybe optional)
pyopengl (only needed for 3d viewing of networks)

There are some actual real-world examples in examples/ in the distribution. Some of those experiments were still being worked on when they were included there, so beware.

Comments and bugfixes are appreciated:
Rich Drewes
drewes at interstice dot com

If there is sufficient interest, an NCS user mailing list and/or Brainlab user mailing list might be started.

page views
local variable 'views' referenced before assignment

Back
Valid HTML 4.0!