/autoencoder-for-lake-bathymetry

Using deep learning to predict lake bathymetry based on the surrounding terrain

Primary LanguageJupyter Notebook

Project: autoencoder-for-lake-bathymetry

Predicting lake bathymetry from the topography of the surrounding terrain using deep learning

Repository containing code, data, model weights, and example notebook for manuscript titled Predicting lake bathymetry from the topography of the surrounding terrain using deep learning published in Limnology and Oceanography: Methods (DOI: 10.1002/lom3.10573).

Python version 3.8 and R version >4.0. See the 'requirements.txt' file for specific Python package versions used for the analysis.

See Example notebook.ipynb for a demonstration of using the model and example data provided in here. Download this GitHub repository as a '.zip', extract the contents, and open the Jupyter Notebook file. To run the notebook, install the following Python packages: torch (deep learning library), rasterio (geographic data handling), and optionally matplotlib for plotting.

Example of ground truth and predicted lake bathymetry