Upload fitted `brms` model to GitHub using Git Large File Storage

bohm0072 / bayesian-cndc-potato

Repository associated with the Bohman et al. (2022) manuscript "Quantifying critical N dilution curves across G × E × M effects for potato using partially-pooled Bayesian hierarchical methods"

0 stars 1 forks source link

Upload fitted `brms` model to GitHub using Git Large File Storage #2

Open bohm0072 opened 3 years ago

bohm0072 commented 3 years ago

Given the fitted brms model of size ~200 MB, determine the optimal method to store this model within this repository using Git Large File Storage for subsequent reference in context of the published manuscript and enabling reproducible analysis without needing to reproduce the model.

To do

[ ] 1. Identify how to implement Git LFS for this particular file and repository
[ ] 2. Upload the fitted, final model to Git LFS

bohm0072 commented 3 years ago

Currently, this issue is being handled using the googledrive library in a workaround fashion where the saved model can be stored and retrieved from the manuscripts/models directory from Google Drive (rather than from GitHub). This works, but a more robust implementation might be needed. Issues here include lack of version control. Upside is no restrictions on storage or bandwidth, for the time being.