This should make it easier for people to modify the datasets. I don't (yet) know how to give Zenodo access to that file to every contributor to mlpack, but perhaps we can figure out that part later.
This change to the website script simply grabs the Zenodo dataset URL instead of grabbing it from my personal website.
I uploaded
datasets.tar.gzto Zenodo like @bkmgit suggested in https://github.com/mlpack/examples/issues/28.https://zenodo.org/record/4042173
This should make it easier for people to modify the datasets. I don't (yet) know how to give Zenodo access to that file to every contributor to mlpack, but perhaps we can figure out that part later.
This change to the website script simply grabs the Zenodo dataset URL instead of grabbing it from my personal website.