[FEA] This repo is too big. Consider using git-lfs or keeping the data outside of git.

[harrism]

Describe the bug
This repo is 500MB and growing. Cloning is slow. Most users probably don't use all the datasets. Should consider externalizing the datasets. Could also use git-lfs (large file storage):
https://git-lfs.github.com

(base) mharris@dgx02:~/rapids$ git clone git@github.com:harrism/notebooks-contrib.git
Cloning into 'notebooks-contrib'...
remote: Enumerating objects: 4547, done.
remote: Total 4547 (delta 0), reused 0 (delta 0), pack-reused 4547
Receiving objects: 100% (4547/4547), 489.83 MiB | 24.91 MiB/s, done.
Resolving deltas: 100% (2392/2392), done.
Checking out files: 100% (213/213), done.

[taureandyernv]

@harrism , keeping the repo size down was a problem, but we should not be at 500mb. IIRC, 300ish. we may have had an accidental push of dataset or history. If josh okays it, i will search and remove the old datasets. thanks man!

[harrism]

@taureandyernv Even if you remove them, they will still be there in the git history (affecting everybody who clones), unless you modify the history. This is why you should use a different system for large datasets, like git-lfs, since it replaces the files with text pointers to external storage on GitHub.