Code for the following publication:
Liesch, T., Ohmer, M. (2025): Strategies for Incorporating Static Features into Global Deep Learning Models (submitted to HESS)
Contact: tanja.liesch@kit.edu
The corresponding dataset is hosted on Zenodo and can be accessed here: https://doi.org/10.5281/zenodo.16601180
All models can be found in the models folder. IS refers to in-sample models, OOS to out-of-sample models. Models without suffix are run with environmental static features, model with the suffix _ts with time series static features. The model names (conc, att...) correspong to the abbreviations introduced in the above mentioned paper.
The evaluation routine used to calculate the error metrics is available in the evaluation folder.
This project is licensed under the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
© 2025 KIT Hydrogeology. Commercial use is not permitted.
