Regular_and_Singular_SPDEBench is a benchmark dataset for evaluating machine learning-based approaches to solving stochastic partial differential equations (SPDEs).
In the SPDEBench, we generate two classes of the datasets: the nonsingular SPDE datasets and a novel singular SPDE dataset.
The nonsingular SPDE contains:
Stochastic Ginzburg-Landau equation(Phi41):
Stochastic Korteweg–De Vries equation(KdV);
Stochastic wave equation;
Stochastic Navier-Stokes equation(NS).
