[R-pkg-devel] Best practice for releasing heavy internal data
Hi all, I am revisiting this and wondering if it would be a case in which I should submit as-is, and include justification for ignoring the NOTE about excessive directory size. Best, Paul Hibbing
On Mon, Feb 18, 2019 at 3:20 PM Paul Hibbing <paulhibbing at gmail.com> wrote:
Hi all, I am preparing a package that makes use of several neural networks, which are quite large. They aren't intended for public use, so they're currently stored in sysdata.R, but they make the R directory too heavy (close to 5MB). What is best practice in this case? I'm aware the general recommendation is to set up a separate data package that updates infrequently, but that seems undesirable since these objects weren't going to be exported in the first place. Thanks for your help. -Paul Hibbing