This repository contains the code used for the papers:

• Bayesian calibration of stochastic agent based model via random forest
• Advancing calibration for stochastic agent-based models in epidemiology with stein variational inference and gaussian process surrogates

It also contains hospitalization and death data produced by the CityCOVID agent based model. As part of these papers, it provides code to train a Random Forest or Gaussian Process surrogate model for CityCOVID hospitalizations and deaths, calculate a Bayesian or Stein variational estimate of parameters from CityCOVID using the surrogate, and then produce plots and data of these estimates.

Package installation for this code is mostly easily accomplished using pixi. All dependencies can be installed by running pixi install in the repository root. However, specifics can be found below.

The code in this repository makes use of the following python packages:

• matplotlib
• numpy
• pandas
• properscoring
• pymcmcstat (from updated version here)
• seaborn
• scikit-learn

The code in this repository makes use of the following R packages:

• ggdist
• coda
• mcgibbsit
• scoringutils

Instructions to reproduce each paper can be found in the papers/**/README.md files. However, generally, the various scripts from scripts/** will be used to call utility functions in src/ using data from data/ and will output files into results/** and plots into plots/**.