Improved Workflow With Isolated Jupyter Environments
summary
A tiny tool called Callisto wires a virtualenv or conda env into JupyterLab so a notebook gets a kernel named for that environment; two commands are all that’s required: pip install callisto and callisto -p $PWD. The demo builds a small primes package, runs tests from a notebook, installs numpy into the env, and claims a Sieve implementation checks primes up to 100 million in ~2s. Works with venv, conda and ipykernel.
AI-written summary · about
chapters (20)
- 0:00Introduction
- 1:01Overview
- 1:46Jupyter Life
- 4:15Callisto
- 5:01New Environment
- 5:19New Notebook
- 5:46Demo
- 6:43iPad Kernel
- 7:35Virtual Environments
- 9:49Conda
- 12:11PiN
- 13:34History
- 16:27Bash aliases
- 17:13Starting a small project
- 18:14Creating a new project
- 19:14Opening Jupyter Lab
- 19:58Primes Library
- 21:31Primes Test
- 21:55Python Path
- 22:30Second Implementation
related talks
36mHow to Roll your own Dockerized Jupyter Environment
01h 18mProductionizing Jupyter Notebooks
22mDock Your Jupyter Notebook
29mPython + VPS JupyterHub/Notebook...
01h 36mDeploying Jupyter Notebooks for students and researchers
24m"Invisible Walls: Isolating Your Python"
39mNotebook To Production
41mTaming the toxic python environment on your laptop