Parametric generation of conditional geological realizations using generative neural networks (arXiv)

Requires PyTorch 0.4+

Run with python main.py [--options]

• Z

• X

• O

You can download our pre-trained unconditional generator netG.pth here (12MB), which has been trained using Wasserstein GAN (https://arxiv.org/abs/1701.07875)

• main.py main code
• models.py neural network architectures
• utils.py helper functions
• dat/cond*.dat conditioning test cases