FloodSformer: River Flood datasets&checkpoints

Data used for the paper: Pianforini et al. (2025). FloodSformer: A transformer-based data-driven model for predicting the 2-D dynamics of fluvial floods. Environmental Modelling & Software. https://doi.org/10.1016/j.envsoft.2025.106599 The repository contains the training/testing datasets and the checkpoints for the Toce River case study. The data for the Po River test case are unavailable due to restrictions on data permissions. The python code of the FloodSformer model is available at the GitHub repository. More detailed information on the repository content are provided in the README_dataset.md file. Acknowledgements This research was granted by University of Parma through the action “Bando di Ateneo 2024 per la ricerca”. RV and SD acknowledge financial support from the PNRR MUR project ECS_00000033_ECOSISTER. This research also benefits from the HPC facility of the University of Parma. Finally, the Authors acknowledge the CINECA award under the ISCRA initiative, for the availability of high-performance computing resources and support (projects AMNERIS and MOZART).