Skip to content

DeepWave-KAUST/Siamese_FWI-pub

Repository files navigation

LOGO

Official reproducible material for SiameseFWI: A Deep Learning Network for Enhanced Full Waveform Inversion - Omar M. Saad, Randy Harsuko, and Tariq Alkhalifah

Project structure

This repository is organized as follows:

  • 📂 asset: folder containing logo;
  • 📂 data: folder containing Marmousi2 and overthrust models data;
  • 📂 utils: set of common function to run FWI;
  • 📂 Model: containing Siamese network;
  • 📂 results: containing the reconstructed velocity model using SiameseFWI;
  • 📂 deepwave-old: containing the old version of the DeepWave package.

Notebooks

The following notebooks are provided:

  • 📙 SiameseFWI_Marmousi.ipynb: the main notebook performing the SiameseFWI for Marmousi model;
  • 📙 SiameseFWI_overethrust.ipynb: the main notebook performing the SiameseFWI for overthrust model;
  • 📙 SiameseFWI_bad_initial_Overthurst.ipynb: the main notebook performing the SiameseFWI for overthrust model; bad initial model;

Getting started 👾 🤖

  • To ensure the reproducibility of the results, we suggest using the FWIGAN.yml file when creating an environment.
  • Please install deepwave 0.0.8 version, which is used in this project.

Run:

./install_env.sh

It will take some time, but if you see the word Done! on your terminal you are ready to go.

Remember to always activate the environment by typing:

conda activate FWIGAN

To install the old version of the DeepWave package, navigate to the "deepwave-old" folder, and run:

cd ./deepwave-old/
python setup.py install

Disclaimer: All experiments have been carried on a Intel(R) Xeon(R) CPU @ 2.10GHz equipped with a single NVIDIA GEForce RTX 3090 GPU. Different environment configurations may be required for different combinations of workstation and GPU.

Cite us

@article{saad2024siamesefwi,
  title={SiameseFWI: A deep learning network for enhanced full waveform inversion},
  author={Saad, Omar M and Harsuko, Randy and Alkhalifah, Tariq},
  journal={Journal of Geophysical Research: Machine Learning and Computation},
  volume={1},
  number={3},
  pages={e2024JH000227},
  year={2024},
  doi={https://doi.org/10.1029/2024JH000227},
  publisher={Wiley Online Library}
}

About

Official reproducible material for SiameseFWI: A Deep Learning Network for Enhanced Full Waveform Inversion

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 2

  •  
  •