Generating Random Data with Epileptor

[canthony....@gmail.com]

Hi all,

I'm currently trying to generate data with the Epileptor model with different random variations across multiple runs. 

I initially tried using Additive noise, but that didn't seem to work. The noise added seemed to be the same in each run.

I then tried updating the noise_seed of the Additive noise with a different value each run. That generated random variations. However, in a number of instances, that also led to the model outputting NaN values.

Are there any suggestions on how I might generate random variations across runs?

Thank you,

Charles

[WOODMAN Michael]

On 1/5/2022 3:16 PM, canthony....@gmail.com wrote:

> I then tried updating the noise_seed of the Additive noise with a
> different value each run. That generated random variations. However, in
> a number of instances, that also led to the model outputting NaN values.

It sounds like you should just lower the level of noise or the time step.