Reproduce “An evaluation of deep learning models for predicting water depth evolution in urban floods” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Stefania Russo, Nathanaël Perraudin, Steven Stalder, Fernando Perez-Cruz, Joao Paulo Leitao, Guillaume Obozinski, Jan Dirk Wegner
The authors train neural networks to predict future urban water-depth maps by learning from CADDIES simulations, using terrain, rainfall, and recent depths. The models predict depth changes at five-minute steps. On short events in catchment 709, U-Net and Graph reduce errors at deeper depths, but show no advantage on tested long events or catchment 744. This does not establish accuracy against observed floods or operational readiness.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.