Bayesian Approach to Experimental Design Applied to Compton Scattering

This repository contains the data and Jupyter notebooks to reproduce and extend the figures in:

• Melendez, Furnstahl, Griesshammer, McGovern, Phillips, Designing Optimal Experiments: An Application to Proton Compton Scattering, arXiv:2004.11307.

Overview

The directory notebooks contains all the relevant Jupyter notebooks, including the main notebooks main_manuscript_analysis.ipynb, plot_compton_coefficients.ipynb, and order_exponent_analysis.ipynb, which generate the figures in the paper and saves them to the subdirectory manuscript_figures. The directory compton contains the Python implementation code. The raw data can be found in data. More information can be found in the annotated notebooks.

Requirements and Installations

Due to the large size of the data files, cloning this repo requires git-lfs. The installation instructions for git-lfs can be found at git-lfs.github.com.

Python 3 is required with the (standard) packages listed in requirements.txt installed. They can be installed by running the command:

pip3 install -r requirements.txt

In addition, J. Melendez's package gsum, which is publicly available here including installation instructions, needs to be installed separately. Do not use gsum as installed by pip3.

With these prerequisites, to install this repository simply run (at the top level):

pip3 install -e .

Contact

To report any issues please use the issue tracker.

Citing this Work and Further Reading

• Melendez, Furnstahl, Griesshammer, McGovern, Phillips, Designing Optimal Experiments: An Application to Proton Compton Scattering, arXiv:2004.11307.