Coder Social home page Coder Social logo

elaloy / vae_mcmc Goto Github PK

View Code? Open in Web Editor NEW
9.0 2.0 4.0 4.27 MB

Inversion with a VAE-based low-dimensional parameterization for complex geologic priors (Python 2.7)

Python 34.38% Visual Basic 65.62%
variational-autoencoders inversion geophysics hydrology mcmc-inversion geology

vae_mcmc's Introduction

Using a VAE-based low-dimensional parameterization to solve inverse problems with complex geologic prior information

This Python 2.7 package contains the (1) convolutional variational autoencoder (VAE) and (2) DREAM(ZS) Markov chain Monte Carlo (MCMC) sampler used for probabilistic inversion of hydrologic data within complex binary geological media by Laloy et al. (2017).

Large data files

Because of Github space limits, neither the training sets (of geologic model realizations) nor the trained VAE models (needed to reproduce the results presented in the paper) are provided herein. But these are of course available. Please drop me an email if you want to get these files.

Building the VAE-based representation

The codes for training our VAEs with Theano/Lasagne are contained in vaecnn2D_train.py (for the two 2D considered case studies) and vaecnn3D_train.py for the two 3D considered case studies). Also, realizations generated by the trained VAE models can be produced by running the model_generation.py script.

Note that since the development of Theano has just stopped, I am planning to rewrite the code in Tensorflow/Keras. Just need time :-).

Performing the inversion

To perform the MCMC inversion within the VAE latent space using DREAM(ZS), use the run_mcmc.py script from the Inversion folder.

Citation

Please make sure to cite our paper if you use any of the contained code in your own projects or publication:

Eric Laloy, Romain Hérault, John Lee, Diederik Jacques, Niklas Linde, Inversion using a new low-dimensional representation of complex binary geological media based on a deep neural network, Advances in Water Resources, 2017, 110:387-405.

License

Licence: the Theano/Lasagne code for the VAE is under MIT license while the MCMC inversion code is under GPL license. See the corresponding folders for details.

Contact

Eric Laloy ([email protected])

vae_mcmc's People

Contributors

elaloy avatar

Stargazers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

Watchers

 avatar  avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.