[Comp-neuro] BindsNET - A Machine Learning Spiking Neural Networks Library in Python

Hananel Hazan hananel at hazan.org.il
Thu Jan 17 04:02:35 CET 2019


 [Apologies for cross-postings]

Dear colleagues,

It is our pleasure to introduce the release of the python package BindsNET
<https://github.com/Hananel-Hazan/bindsnet>and its companion paper in
Frontiers in Bioinformatics “BindsNET: A Machine Learning-Oriented Spiking
Neural Networks Library in Python
<http://journal.frontiersin.org/article/10.3389/fninf.2018.00089>”.

BindsNET is a python package to simulate a spiking neuronal network using a
GPU. The framework is easy and well in line with Python, NumPy and PyTorch
standard that runs on CPU’s/GPU’s. BindsNet will serve both the novel
researcher and the seasoned programmer wanting to perform rapid prototyping
prior to diving into hardware implementation.

To achieve maximum performance and flexibility, BindsNET re-purposes the
powerful and flexible PyTorch <http://pytorch.org/> library. By wrapping
around PyTorch we avoid "reinventing the wheel" by reusing it's function
and sub-modules for the spiking neuronal networks computational needs.

For more details on performance, usability and code examples please see the
Frontier paper
<http://journal.frontiersin.org/article/10.3389/fninf.2018.00089> as well
in the GitHub repository <https://github.com/Hananel-Hazan/bindsnet>.

BindsNet welcomes researcher, students and neuro-scientists to both utilize
and contribute to this leading-edge toolbox.

*Happy spiking*


Thanks

Hananel Hazan
Postdoctoral Research Associate
BINDS Lab, UMass Amherst, USA
http://Hananel.Hazan.org.il
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.tnb.ua.ac.be/pipermail/comp-neuro/attachments/20190116/9fd95c74/attachment.html>


More information about the Comp-neuro mailing list