Python

Pip

• update Pip version using pip itself

pip install pip -U

Advanced

Virtual Environments

Package dependencies are sometimes incompatible to each other, so the solution is to create a isolated package tree shell to have a "clean" system for every package installation. In python it's the Virutal Environment

• create:

python3 -m venv new-env

• activate

source ~/new-env/bin/activate

JupyterHub

Installation

• create a user (jupyterhub

python3 -m venv jupyterhub-env
. jupyterhub-env/bin/activate 
(jupyterhub-env) ~ $ pip install jupyterhub

• Extension IPython (using JupyterLab 2)

pip install ipympl
pip install nodejs
jupyter labextension install @jupyter-widgets/jupyterlab-manager
jupyter labextension install jupyter-matplotlib

Config

• Users

adduser -G jupyterhub newuser

• in /etc/sudoers

Cmnd_Alias JUPYTER_CMD = /usr/local/bin/sudospawner
rhea ALL=(%jupyterhub) NOPASSWD:JUPYTER_CMD

Use JupyterLab by Default (instead of Notebook) in /etc/jupyterhub/jupyterhub_config.py

c.Spawner.cmd=["jupyter-labhub"]

• change default notebook directory

   c.Spanwer.default_url: /notebooks
   c.Spawner.notebook_dir: /apps/jupyter

• Running as root is not recommended. Use --allow-root to bypass.

c.Spawner.args = ['--allow-root']

Debug

• run with debug from cli

jupyterhub -f /etc/jupyterhub/jupyterhub_config.py --debug

• [D 2022-02-15 10:27:43.776 JupyterHub app:2727] Could not load pycurl: No module named 'pycurl'

pip install pycurl