taylorjwright / r_and_p_versioning Goto Github PK

Merge the targets branch into main and cleanup + documentation

Compilation of the paper on Github Actions is a bit hacky for now, need to simplify it

Installing the dependencies as they were around the time the paper was written is proving to be quite difficult. This is because version 4.9-6 of {Matching}, which was current around december 2019 was, for some reason, not archived.

I'll try to install it from github directly, but another solution would be to not care about the version of packages and only focus on the R version.

recode() is not very robust and seems to have changed since this paper was published. Need to use case_when()

just to make sure: the outputs we should compare between R 3.5 and R 3.6 are the estimated models from Zeilig and ZeiligChoice, right? The results are different due to how R computes random numbers.

Just want to make sure

I have the visualisations of the networks to replicate table 1.

As of now, R 3.5.2 gets pulled. Somehow there's an issue with my Docker installation such that pulling 3.5.0 doesn't work, but I'll fix it by building and pushing the image on Dockerhub using Github Actions