This is a project which is currently making use of HPC facilities at Newcastle University. It is active.
For further information about this project, please contact:
The project combines physically based flood simulation outputs, including CityCAT-style hydrodynamic data, with machine learning and deep learning methods such as DNO/TL-DNO, SWE-GNN and U-RNN. The aim is to develop and test workflows for faster flood prediction, flood depth/velocity estimation, and model generalisation to data-scarce catchments.
Python; PyTorch; CUDA; PyTorch Geometric; torch-scatter; torch-sparse; torch-cluster; torch-spline-conv; NumPy; SciPy; pandas; scikit-learn; xarray; h5py; netCDF4; GDAL; rasterio; rioxarray; geopandas; shapely; pyproj; fiona; matplotlib; JupyterLab; Git; VS Code Remote/SSH; conda/mamba or micromamba; and potentially Apptainer/Singularity containers for reproducible environments.
Research repositories/workflows include DNO/TL-DNO, SWE-GNN/mSWE-GNN, U-RNN/URNN and flood data preprocessing pipelines.