Groundwater Level Forecasting with ANNs: A Comparison of LSTM CNN and NARX

doi of this repo:

doi of according publication: https://doi.org/10.5194/hess-25-1671-202

This repository enables you to reproduce the results and apply the groundwater level forecasting methodology of:
Wunsch, A., Liesch, T., Broda, S., Groundwater level forecasting with artificial neural networks: a comparison of long short-term memory (LSTM), convolutional neural networks (CNNs), and non-linear autoregressive networks with exogenous input (NARX)

Contact: andreas.wunsch@kit.edu

ORCIDs of authors:
A. Wunsch: 0000-0002-0585-9549
T. Liesch: 0000-0001-8648-5333
S. Broda: 0000-0001-6858-6368

For a detailed description please refer to the publication. Please adapt all absolute loading/saving and software paths within the scripts to make them running, you need Matlab and Python software for a successful application. We further use the Python Package BayesianOptimization by fmfn. To run the Python Code please download and install this package.

Content Overview:

• /CNN - Python Code
  Contains Python scripts of the models and necessary example files.

• /LSTM - Python Code
  Contains Python scripts of the models and necessary example files.

• /NARX - Matlab Code
  Contains Matlab scripts of the models and necessary example files.