@claraqin I suggest having two folders: data-raw and data. data-raw stores the data directly downloaded from NEON, etc. using code, and it is a symbolic link on the server and won't be uploaded to GitHub. data stores processed data that are needed to run the models, and should be relatively small and will be version-controlled on GitHub. For example, the rda files can be put in the data folder. That way, collaborators only need the GitHub repo without accessing the server.
@claraqin I suggest having two folders:
data-raw
anddata
.data-raw
stores the data directly downloaded from NEON, etc. using code, and it is a symbolic link on the server and won't be uploaded to GitHub.data
stores processed data that are needed to run the models, and should be relatively small and will be version-controlled on GitHub. For example, therda
files can be put in thedata
folder. That way, collaborators only need the GitHub repo without accessing the server.