Regularization from examples via neural networks for parametric inverse problems: topology matters

• Generate the datasets of loop parameters and corresponding visibilities by running "data_generation.ipynb";
• Train the networks with "nn_training_simple_dataset.ipynb" and "nn_training_complete_dataset.ipynb";
• Plot the results with "plot_results_simple_dataset.ipynb" and "plot_results_complete_dataset.ipynb"