This repository contains material for an uncertainty quantification (UQ) class on equation of state (EOS) that was presented at Science of Compression in Condensed Matter (SCCM) 2025. It includes a Python package called eosuq for analyzing linear shock compression data. The package performs Bayesian linear regression analytically and with Markov Chain Monte Carlo, and provides a bootstrap approach for comparison. The methods are demonstrated in a notebook on publicly available data contained in the notebooks directory. The dataset, from shock compression experiments on copper, is from pages 57-60 of Marsh, S. P. (1980), "LASL Shock Hugoniot Data".
- Clone the repository
git clone https://github.com/LLNL/SHOCK-UQ.git
cd SHOCK-UQ- Create a virtual environment
python3 -m venv .venv- Activate the virtual environment
source .venv/bin/activate- Install dependencies
pip install -r requirements.txt- Install the
eosuqpackage locally
pip install -e .- Launch Jupyter notebook The notebook is located in the
notebooksdirectory.
- Jason Bernstein (Lawrence Livermore National Laboratory)
- Justin Brown (Sandia National Laboratories)
- Beth Lindquist (Los Alamos National Laboratory)
This software is distributed under the terms of the MIT license. All new contributions must be made under the MIT license.
See LICENSE and NOTICE for details.
LLNL-CODE-2005336
