Jupyter correct workflow

https://samedwardes.com/blog/2022-10-23-best-jupyter-lab-install/ reference

Jupyter Basic Environment Creation and Customization

Part 1: Making a Python Kernel Available in Jupyter (Installed via pipx)

1. Install Jupyter via pipx (Already done):

   bash

   pipx install jupyter --include-deps
   pipx ensurepath

   alternative:
   pipx install pretzelai
   pipx ensurepath

2. Set Up a Python Virtual Environment for the Kernel:

   bash

   python3.12 -m venv ~/venvs/python3.12

3. Activate the Virtual Environment:

   bash

   source ~/venvs/python3.12/bin/activate

4. Install ipykernel in the Virtual Environment:

   bash

   pip install ipykernel

5. Add the Kernel to Jupyter:

   bash

   python -m ipykernel install --user --name python3.12 --display-name "Python 3.12"

6. Deactivate the Virtual Environment:

   bash

   deactivate

Part 2: Customizing JupyterLab for a Better Workflow (With pipx)

1. Install JupyterLab Extensions with Pipx

bash

pipx install jupyter --include-deps
pipx ensurepath
pipx runpip jupyter install jupyterlab-code-formatter
pipx runpip jupyter install black
pipx runpip jupyter install jupyterlab-lsp
pipx runpip jupyter install python-lsp-server[all]
pipx runpip jupyter install lckr-jupyterlab-variableinspector
pipx runpip jupyter install jupyterlab-spellchecker
pipx runpip jupyter install jupyter-ai[all]
pipx runpip jupyter install jupyter_ai_magics
pipx runpip jupyter install mitoinstaller
pipx runpip jupyter install jupyterlab-snippets

pipx run jupyter lab build

Other Tips

in order to use %%ai magic command, you will need to install in separate kernel this:
pip install jupyter_ai_magics

another option is to manually install JupyAI

also codeium auto code completion is availabe from codeium

jupyterlab-mutableai can change jupyter code to production code

install jupyterlab-execute-time which can show execution time

in view tab, check the line number

jupyterlabcodetoc can be useful

Pretzel Commands

bash

pipx runpip pretzelai install jupyter
pipx runpip pretzelai install jupyterlab-code-formatter
pipx runpip pretzelai install jupyterlab-lsp
pipx runpip pretzelai install python-lsp-server[all]
pipx runpip pretzelai install black
pipx runpip pretzelai install lckr-jupyterlab-variableinspector
pipx run jupyter lab build

2. Using the Built-in Terminal and File Explorer

JupyterLab comes with built-in support for a terminal and file explorer, allowing you to manage files and run shell commands directly within the interface.

• Open a terminal: Go to File > New > Terminal to open a terminal inside JupyterLab.
• Use the file explorer: The file browser on the left sidebar can be used to open, move, rename, or delete files directly within the JupyterLab environment.

Important Notes:

• Rebuild JupyterLab after installing extensions:
  bash

  pipx run jupyter lab build

• Verify extensions:
  bash

  pipx run jupyter labextension list

  3. Enable an Extension

  bash

  pipx run jupyter labextension enable @jupyterlab/notebook-extension:language-server

disable:

bash

pipx run jupyter labextension disable @jupyterlab/notebook-extension:language-server

Jupyter Kernel Management

Kernel Specification (kernelspec) System

Kernel Spec Locations

Check kernel spec locations:

bash

jupyter kernelspec list

How Kernels Are Registered

bash

python -m ipykernel install --user --name mypython --display-name "Python 3.12"

Remove an Old Kernel

bash

jupyter kernelspec uninstall old-env

Example: Adding Python 3.12 as a Kernel to Pipx Jupyter

1. Create a Virtual Environment for Python 3.12:

   bash

   python3.12 -m venv ~/venvs/python3.12

2. Activate the Environment:

   bash

   source ~/venvs/python3.12/bin/activate

3. Install ipykernel in the Environment:

   bash

   pip install ipykernel

4. Add the Kernel to Jupyter:

   bash

   python -m ipykernel install --user --name python3.12 --display-name "Python 3.12"