Event-SNN-Resources

Code and Examples

The repository provides a collection of Jupyter notebooks that demonstrate the implementation of various techniques and methods related to Event-based Spiking Neural Networks. Below are links to the key notebooks and their descriptions:

Basic Examples

• Basic Event Processing: This notebook covers the fundamental steps for processing event-based data.

Advanced Examples

• Build Feedforward-SNN: A tutorial on building Feed-Forward Spiking Neural Networks using PyTorch.
• Build S-CNN: A tutorial on building Spiking Convolutional Neural Networks using PyTorch.

Techniques and Methods

• Image to Poisson Spikes: This repository includes code to convert images into Poisson spikes.
• Image to TTFS Spikes: Instructions and code to convert images into Time-to-First-Spikes activations.

Datasets

The repository utilizes a variety of datasets. Besides classic RGB datasets, event-based datasets are provided by Tonic, a library for event-based data, which simplifies the handling and processing of event-based datasets. Below are some of the event datasets provided by Tonic:

• N-MNIST
• N-CALTECH101
• CIFAR10-DVS
• IBM DVS Gestures
• POKER-DVS

For more information, refer to the Tonic Documentation and the list of available Tonic Datasets.

SNN Libraries

• NEST: NEST Website
• NEURON: NEURON Website
• Brian: Brian2 GitHub
• Nengo: Nengo Website
• SNNToolbox: SNNToolbox GitHub
• BindsNET: BindsNET GitHub
• snnTorch: snnTorch GitHub
• SpikingJelly: SpikingJelly GitHub
• NxTF: NxTF GitHub

How to Use This Repository

1. Clone the Repository:

   git clone https://github.com/radlab-sketch/Event-SNN-Resources.git