surrogate-pt-Bayeslands

Surrogate assisted parallel tempering Bayeslands

code

We have two major versions

• Surrogate parallel tempering using random-walk proposal distribution for Badlands landscape evolution model: [surrogate_ptBayeslands.py] to run with [run.sh]
• paper online: Surrogate-assisted Bayesian inversion for landscape and basin evolution models

Prerequisites

The framework is built using:

• Parallel tempering Bayeslands
• R. Chandra et. al, Multicore Parallel Tempering Bayeslands for Basin and Landscape Evolution with Arxiv open access
• Badlands model

Installation

• install conda python 3.x version: https://www.anaconda.com/distribution/

installation - linux terminal commands - in conda evironment

• conda install -c plotly chart-studio (install in conda env)
• curl "https://bootstrap.pypa.io/get-pip.py" -o "get-pip.py" (install pip3 in conda env)
• python3 get-pip.py --user
• pip3 install tribad
• pip3 install gFlex
• pip3 install badlands/ (do sudo if needed)

run

• run_uplift.sh
• results in Examples/australia folder

Experiments

• Results with data for surrogate evaluation offline
• Results for landscape evolution problems - selected runs

Versioning

• TBA

Authors

• R. Chandra, D. Azam , A. Kapoor and R. D. Muller Surrogate-assisted Bayesian inversion for landscape and basin evolution models

License

• This project is licensed under the MIT License - see the Open Source Licence file for details

Acknowledgments

• Dr. Nathanial Butterworth, Sydney Informatics Hub, University of Sydney
• Mr. Konark Jain, Indian Institute of Technology

Contact

• Dr. Rohitash Chandra, University of New South Wales (c.rohitash at gmail.com or rohitash.chandra at unsw.edu.au)