The definitive guide to setup my Python workspace
Henrique Bastos

I know this is an old article but I was wondering why split jupyter into a separate virtualenv when the kernel used would require packages installed that is a combined list of all packages in all virtualenvs that want to use jupyter … example: i wanted to use pandas but pandas wasn’t installed in the jupyter space but was in my local virtualenv so I had to re-run

python -m ipykernel install --user

from within the virtualenv that had pandas installed. I suppose I could have installed pandas in the jupyter virtualenv but then I got to thinking, I’d essentially be treating the jupyter virtualenv as a global space of packages and also mirroring them from ALL my other virtualenvs …


Caveat: I am new to this so if I’m missing something stupid let me know :)

One clap, two clap, three clap, forty?

By clapping more or less, you can signal to us which stories really stand out.