To reproduce the numerical results of each model problem, go to the corresponding folder and follow these steps:

1. Execute Main.py to run the code and train the network. The raw-format results are saved in Saved_models.
2. Run Main_fem.py to execute the FEM code and generate approximate FEM solutions used to compare with the LS-ReCoNN results. This step is necessary because an analytical solution is not available for this problem, so the FEM solution serves as the reference for comparison.
3. Run plot1.py and plot2.py for 1D and 2D problems, respectively, to generate the final graphics.