Closed MatthewGrim closed 3 years ago
It turns out this approach has already been explored in: https://www.researchgate.net/publication/337844502_Title_Machine_learning_approaches_for_the_solution_of_the_Riemann_problem_in_fluid_dynamics_a_case_study
For that reason I am closing this issue.
This is an experiment to see if such an approach can even be done. The aim would be to recover a HLLC riemann solver using machine learning methods on data generated from an exact riemann solver. I already have the exact method to generate the necessary data, but need to figure out a way to correctly fit the state.