The two datasets are actually the same.

[amiralizadeh1]

Hi omniscientoctopus,

Thank you for this great implementation.

I tried to use your dataset for an inverse problem on Burger's equation. they look exactly the same when I plot them. I guess there might be a difference in the solution of the PDE once we use different values for nu.

Secondly, In my inverse problem, I tried to identify the parameter nu of the PDE as well as the solutions of the PDE. Unfortunately, the NN can't converge towards the desired value, say nu=0.01/pi.

looking forward to your feedback, best.

[omniscientoctopus]

Hi there, thanks for your question. I did a quick plot and the datasets look different. Difference between $\mu=0.1/\pi$ and $\mu=0.05/\pi$ is a bit subtle though.

b_shock_mu_01_pi = scipy.io.loadmat('Data/burgers_shock_mu_01_pi.mat')
solutionplot(b_shock_mu_01_pi['usol'], 1, 1)

b_shock_mu_005_pi = scipy.io.loadmat('Data/burgers_shock_mu_005_pi.mat')
solutionplot(b_shock_mu_005_pi['usol'], 1, 1)

b_shock_IC_sin2pi = scipy.io.loadmat('Data/burgers_shock_IC_sin2pi.mat')
solutionplot(b_shock_IC_sin2pi['usol'], 1, 1)

Does this answer your question?