cff-version: 1.2.0
abstract: "<p>FluvGAN is a set of Python scripts to test generating and inverting fluvial deposits using generative adversarial networks (GANs) in the context of predicting the physical properties of the subsurface. It builds upon FluvDepoSet (https://doi.org/10.25919/4fyq-q291) - a dataset of 3D representations of fluvial deposits simulated with CHILD (https://github.com/childmodel/child), a landscape evolution model - for training, and upon voxgan (https://github.com/grongier/voxgan), a Python package making it easier to define and train GANs with PyTorch.</p><p><br></p><p><strong>Installation</strong></p><p>You can install all the packages needed to run all the scripts using conda and the file<em> environment.yml</em> included in this repository:</p><p>				conda env create -f environment.yml</p><p>In addition, the file <em>environment_cluster.yml </em>contains the full list of packages with which the scripts were run on a cluster with a Linux system:</p><p>				conda env create -f environment_cluster.yml</p><p>This environment misses some packages to visualize the results, because those steps were not done on the cluster, and it misses voxgan, because it was used during voxgan&#x27;s development (now you can install it using pip, see https://github.com/grongier/voxgan).</p><p><br></p><p><strong>Usage</strong></p><p>FluvGAN is subdivided into three compressed folders:</p><ul><li><em>scripts.zip</em> contains all the Python scripts and Jupyter notebooks to train and test the GANs and to generate some figures showcasing the results.</li><li><em>data.zip</em> contains the well &amp; seismic data used for testing GAN inversion; the training and testing samples need to be downloaded from FluvDepoSet at https://doi.org/10.25919/4fyq-q291, which can be done using the script <em>fluvgan_0_data.py</em>.</li><li><em>outputs.zip</em> contains all the outputs from all the scripts, including the pretrained GAN models; the outputs&#x27; names follow the same convention as the scripts, and the scripts showcase how to reuse those outputs using Python.</li></ul><p>You can find more information in the file <em>README.md</em>.</p><p><br></p><p><strong>License</strong></p><p>Copyright notice: Technische Universiteit Delft hereby disclaims all copyright interest in the program fluvgan written by the Author(s). Prof.dr.ir. S.G.J. Aarninkhof, Dean of the Faculty of Civil Engineering and Geosciences</p><p>© 2025, Guillaume Rongier, Luk Peeters </p><p>This work is licensed under a MIT OSS licence, see the file <em>LICENSE</em> for more information.</p>"
authors:
  - family-names: Rongier
    given-names: Guillaume
    orcid: "https://orcid.org/0000-0002-5910-6868"
  - family-names: Peeters
    given-names: Luk
    orcid: "https://orcid.org/0000-0002-1776-3173"
title: "FluvGAN: Python scripts to test generating and inverting fluvial deposits using GANs"
keywords:
version: 1
identifiers:
  - type: doi
    value: 10.4121/3469c879-f443-4fcf-83a2-b6df56e96714.v1
license: MIT
date-released: 2025-10-15