Quantifying properties of hot and dense QCD matter through systematic model-to-data comparison
TeX Python
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
data
fig
mcmc
.gitignore
README.md
mtd.bib
mtd.tex

README.md

Quantifying properties of hot and dense QCD matter through systematic model-to-data comparison

arXiv:1502.00339 [nucl-th]

This repository contains all relevant code and data for the above paper. In principle, anyone can completely reproduce the results:

  1. Requirements: Python 3 with numpy, scipy, matplotlib, emcee, george, and my custom library mtd.
  2. Run the preprocess scripts in data/exp and data/model.
  3. Run the emulator+MCMC analysis. Unless you have a very fast CPU, a lot of memory, and scipy compiled with Intel MKL, it will probably take prohibitively long to reproduce the level of statistics in the paper. Edit mcmc/train-and-calibrate on line 163 to reduce the number of MCMC steps; try nwalkers=100, nsteps=1000, nburnsteps=500. Then run ./train-and-calibrate glb to perform the analysis for the Glauber model and once again ./train-and-calibrate kln for KLN.
  4. Generate plots and tables with make-plots in fig.
  5. Compile the PDF normally.

Alternatively, feel free to contact me for the original high-statistics MCMC data.