SpykeTorch

by miladmozafari

miladmozafari /SpykeTorch

High-speed simulator of convolutional spiking neural networks with at most one spike per neuron.

144 Stars 42 Forks Last release: Not found GNU General Public License v3.0 39 Commits 0 Releases

Available items

No Items, yet!

The developer of this repository has not created any items for sale yet. Need a bug fixed? Help with integration? A different license? Create a request here:

SpykeTorch

High-speed simulator of convolutional spiking neural networks with at most one spike per neuron. alt text

SpykeTorch is a PyTorch-based simulator of convolutional spiking neural networks, in which the neurons emit at most one spike per stimulus. SpykeTorch supports STDP and Reward-modulated STDP learning rules. The current code is the early object oriented version of this simulator and you can find the documentation in docs folder in PDF format or in our lab website (http://cnrl.ut.ac.ir/SpykeTorch/doc/) in HTML format. Since SpykeTorch is fully compatible with PyTorch, you can easily use it if you know PyTorch. A tutorial is available in the paper titled "SpykeTorch: Efficient Simulation of Convolutional Spiking Neural Networks with at most one Spike per Neuron" which introduces the SpykeTorch package (https://www.frontiersin.org/articles/10.3389/fnins.2019.00625/full).

IMPORTANT: Current version of SpykeTorch does not support negative synaptic weights.

Scripts info: -

MozafariShallow.py
: Reimplementation of the paper "First-Spike-Based Visual Categorization Using Reward-Modulated STDP" (https://ieeexplore.ieee.org/document/8356226/). -
MozafariDeep.py
: Reimplementation of the paper "Bio-Inspired Digit Recognition Using Reward-Modulated Spike-Timing-Dependent Plasticity in Deep Convolutional Networks" (https://www.sciencedirect.com/science/article/abs/pii/S0031320319301906). -
KheradpishehDeep.py
: Reimplementation of the paper "STDP-based spiking deep convolutional neural networks for object recognition" (https://www.sciencedirect.com/science/article/pii/S0893608017302903). -
tutorial.ipynb
: A brief tutorial on designing, training, and evaluating a SNN with SpykeTorch.

We use cookies. If you continue to browse the site, you agree to the use of cookies. For more information on our use of cookies please see our Privacy Policy.