This folder contains a tensorflow implementation for the DCN and RPN architectures of the paper:

• Agon Serifi, Tobias Günther, Nikolina Ban
  Spatio-Temporal Downscaling of Climate Data using Convolutional and Error-Predicting Neural Networks

The file model.py contains functions to construct the proposed architectures.
Use 'get_model(residual = False)' for the DCN and 'get_model(residual = True)' for the RPN architecture.

• Tensorflow

If you find our work useful to your research, please consider citing:

@article{serifi2021spatio,
  title={Spatio-Temporal Downscaling of Climate Data using Convolutional and Error-Predicting Neural Networks},
  author={Serifi, Agon and G{\"u}nther, Tobias and Ban, Nikolina},
  journal={Frontiers in Climate},
  volume={3},
  pages={26},
  year={2021},
  publisher={Frontiers}
}

By downloading and using the code you agree to the terms in the LICENSE.