Parameters used in generating the Burgers dataset you are sharing

[federicobergamin]

Hi,

I need to generate longer Burger trajectories. I think found the MATLAB files to solve the Burgers equation (I assume you follow this code https://github.com/PredictiveIntelligenceLab/Physics-informed-DeepONets/tree/main/Burger/Data). Can you please share which parameters you used to solve the trajectories in your public dataset?

Can you confirm these are the parameters used for the Gaussian random field?
gamma = 4;
tau = 5;
sigma = 625;

However, I am not quite sure about the time and space discretization. Did you generate the data using the following values
s = 1024;
steps = 1000;
nn = 1024;
and then subsampled to get trajectories with 101 time-steps and 128 spatial discretization? Or which parameters did you use?

Thanks for the help!