The package enables the development of deep learning based surrogates for semi-distributed and distributed hydrological models.
git clone https://github.com/interTwin-eu/hython.git
cd ./hython
pip install .
The goal of the demo is to show how to train and evaluate the models offered in hython.
Every package needed to run the notebooks can be installed by running:
cd ./hython
pip install .[complete]
There are two notebooks showcasing the training of a simple LSTM model and a Convolutional LSTM.
./demo/train_lstm.ipynb
./demo/train_convlstm.ipynb
Todo: evaluation notebooks
python preprocess.py --config preprocessing.yaml
python train.py --config training.yaml
python evaluate.py --config evaluating.yaml
Please open an issue if you have a bug, feature request or have an idea to improve the package.
Class diagram
For further information please contact: