Spike Generation from Input Analog Data in Nengo Implementation


I am new to Nengo Spiking Network implementations and am trying to understand the operation. I am currently running the optimized spiking neural network implementation on MNIST dataset that is given at: https://www.nengo.ai/nengo-dl/examples/spiking-mnist.html.

I wanted to know exactly how the input pixel data is being converted to spikes and which portion of the code or library function is responsible for the same.

Hello tkolanka,

The input pixel data isn’t converted into spikes in a pre-processing step, it is fed directly into the first layer of neurons, as a continuous input value (and then the output of those neurons will be spikes).