How to best create, maintain and archive custom environments from within Jupyter? .. just updated the documentation for Carto-Lab Docker with examples for Python [1] and R [2].

The tricky part is linking Kernels from custom envs with a Jupyter kernelspec (specifically if the Jupyter server and the Kernel are in two different environments). However, most of this can be stored in Jupyter notebook cells, for reproducibility.

There's also a section on archival of package versions with Conda's `env export` (yml approach) and `conda list --explicit` (full archival).

[1]: https://cartolab.theplink.org/use-cases/#create-your-own-environment-in-a-bind-mount-and-install-the-ipkernel
[2]: https://cartolab.theplink.org/use-cases/#example-create-an-environment-with-a-specific-r-version

#jupyter #R #ipython # #reproducibility #Notebooks #conda