Sources for: Avoiding tipping points in fisheries management through Gaussian Process Dynamic Programming

This directory contains the .Rmd files for the manuscript and supplement. These files use functions from the R package from this repository (nonparametric-bayes) but also several additional R packages and other software which are responsible for the parametric comparisons, as well as software responsible merely for the formatting to go from .Rmd to a .pdf file (e.g. knitr, pandoc, LaTeX).

All of these additional dependencies means that rebuilding the manuscript from scratch means more software to install and more points of failure. To help you get started, we provide a copy of the software environment we use in this paper as a Docker container. If you'd rather not worry about the installation details, Install Docker⁠ on your laptop or a user-friendly cloud server⁠ [^1]

The badge above is automatically generated when the manuscript is rebuilt on a remote server. This provides a convenient way to confirm that everything is still working.

Docker Quickstart

Install Docker⁠ on your laptop or server. Then from the manuscripts/ directory run:

docker build -t cboettig/nonparametric-bayes .

docker run --rm -ti -v $(pwd):/data cboettig/nonparametric-bayes R -e "rmarkdown::render('manuscript.Rmd')"

You can now open manuscript.Rmd or supplement.Rmd files in RStudio and run the code interactively or compile the pdfs from scratch. See RStudio's rmarkdown⁠ for details.

Local Installation requirements

• RStudio (includes pandoc)
• A LaTeX environment
• The nonparametric-bayes package from github⁠, including all packages in the DESCRIPTION file (by default suggested packages are not installed).

[1]: this link provides $10 credit to Digital Ocean; enough to run a 1GB instance for a month. Disclosure: If you subsequently become a Digital Ocean customer the authors receive a credit for the referral.