@davidbrochart Intereresting. Does Jupyter need to be in the isolated environment or the kernel would be enough? I foresee the environment creation as very useful, but the launching server mechanism could prevent this project from working with JupyterHub. Could this a jlab ext instead?
@yakshavers Cool! Thanks for sharing! FYI, there is a JupyterHub in HPC monthly meeting. Next one is tomorrow if you want to drop by and share some more - https://t.co/b3cHVd2Ktk
I have a new work time-boxing method. It's called the PomoDOOMo method. I set a timer and work for 45 minutes, and when the timer is up, I check Twitter to see if WW3 has started.
We're happy to share that our paper "EESSI: A cross-platform ready-to-use optimised scientific software stack" has been published in the "Software: Practice and Experience" special issue on "New Trends in #HPC".
Go check out the paper (open access!) at https://t.co/vuqJHYmERm
@betatim Sympy's sympyfy and lambdify were my original source of inspiration to write deap's compile function that wraps arbitrary code in a lambda function. It uses eval and which is easier than AST.
Ref:
https://t.co/5zHVqeOiRO
https://t.co/bbyoudWuqI
One of the hardest and also most useful approaches to improving the open source ecosystem is to commit to using tools that you know won't do exactly what you want, in order to incentivize yourself to contribute and improve them.
New podcast out with @kelseyhightower sharing tales from the first GopherCon and KubeCon, the backstory behind Kubernetes The Hard Way, and how thinking of eng orgs like restaurants can help you choose which dev and infra tools to use. Listen here! https://t.co/8LW9bRDoic