mSWE-GNN (Repository for paper "Multi-scale hydraulic graph neural networks for flood modelling")

(Version 1.1 - Nov. 28th, 2024)

Overview

For reproducing the paper's results, explore plot_results.ipynb

For training the model run main.py

For training and exploring the model, run main.ipynb

For testing a model, run test_model.py

Both main.py and main.ipynb use a config.yaml as reference configuration file.

The repository is divided in the following folders:

database: Creation of hydrodynamic simulations (D-Hydro simulations.ipynb) [requires the license and installation of "D-HYDRO Suite 1D2D"] and conversion of the NETCDF output files into PyTorch Geometric-friendly data (create_dataset.ipynb). Also contains the output of the hydrodynamic simulations (raw_datasets: for downloading the datasets go to https://doi.org/10.5281/zenodo.13326595). This is converted into Pickle files that are then stored and separated into training and testing datasets in datasets.
models: Deep learning models developed for surrogating the hydraulic one: contains a base class with common inputs and functions and one for the SWE-GNN and mSWE-GNN models.
results: Contains results and trained models of the mSWE-GNN Pareto front, used for the paper's results.
training: Contains loss and training functions.
utils: Contains Python functions for loading, creating and scaling the dataset. There are also other miscellaneous functions and visualization functions.

The required libraries are in requirements.txt.