Creating a Package
There are many ways to create a package for MLHub. Indeed, a design goal of MLHub is to simplify the sharing of the work that we do. If you want to create an R based package, then follow the examples like animate or rain. For a Python based package follow the examples like colorize.
The packages can live on gitlab, github, or bitbucket.
If the starting point is a Jupyter Notebook, use the mlhub utility, ipynb2mlhub, which will wrap the text of the notebook within print statements and the code blocks as executable code within a single script file.