A packaging ecosystem is great when it contains all the stuff you need. Here is a simple way to add binary releases made by projects on @github.com:
prefix.dev/blog/octoco...
The next scikit-learn release will allow inspecting the type and values of attributes of fitted estimators in Jupyter notebooks & example code rendered as HTML in sphinx-gallery powered project websites.
scikit-learn.org/dev/auto_exa...
I think the canonical argument is by Dijkstra:
www.cs.utexas.edu/~EWD/transcr...
In the specific context of Python there is:
python-history.blogspot.com/2013/10/why-...
I have added a new tutorial on discrete diffusion models:
github.com/gpeyre/ot4ml
A great overview of the evolution of #container technology via #Docker. It also provides a quick peek into the various technology bits that enable containers work.
If you want to learn more about how containers work, then consider starting with wizardzines.com/zines/contai... by @b0rk.jvns.ca.
Pixi skills are kinda cool. A couple extremely useful ones already hosted on prefix – such as the `conda-forge` skill, or the `rattler-build` skill. It's useful because it contains a bunch of expert knowledge, digestible for agents.
We gave Claude a 15-page PyMC best-practices cheat sheet.
Pass rates jumped from 60% → 93%.
Hardest tasks saw the biggest gains:
• Stochastic volatility: 0% → 67%
• Runtime: 19 min → <3 min
and more ..
👉Benchmark: dub.link/KZxxZfQ
👉 Get the skill: hub.decision.ai/skills/pymc-...
#BayesianModeling
At @quantstack.bsky.social we designed novel bit-unpacking SIMD optimizations for @arrow.apache.org and #ApacheParquet, and implemented them entirely using C++ metaprogramming instead of Python-based code generation.
We'll publish a deep dive blog post soon.
github.com/apache/arrow...
Seems like it might be useful for easier code navigation in FastAPI apps.
🔖
notebook.link combines the power of @jupyter.org's JupyterLite, mambajs and jupyterlite-ai to bring to you the best of open-source data analysis and visualization tools, all within your browser: no installation required! 🤘
#DataScience #OpenSource #Jupyter #WebAssembly #AI #Python
🎯
Just published textual-editor to PyPI.
It's a textual widget that is similar to a TextArea, but that actually runs a real editor and connects it to an embedded terminal emulator, so that you get all the power of neovim (or whatever your preferred editor is).
github.com/davidbrochar...
When you're building recipes for conda-forge, there is an example list that you can use as inspiration for your own recipe!
Hopefully we can extend this list with more community help!
conda-forge.org/docs/mainta...
Stoked to share that the Steering Council has accepted PEP 814. frozendict is coming to Python 3.15!
discuss.python.org/t/pep-814-ad...
ok so newsletter is out, and i've finally posted my PyCon UK keynote with additional material to my homepage.
if you wanna know what took me so long: check out my latest newsletter ;)
back to sauna!
hynek.me/talks/python-superpo...
As a modern reimplementation of Jupyter server, I hope `jupyverse` gains traction, and I think this will really help with that! 🚀
@github.com - you need to get on top of this. It's a big problem and will only get worse if you don't give maintainers the tools to deal with it.
Don't let GitHub descend into a AI/bot cesspit.
Identify AI accounts and give maintainers the ability to block.
Everybody is talking about secure Python sandboxes for LLM code execution, but what about using the browser sandbox?
Quick demo using JupyterLite and the Pyodide kernel 💡
When they said AI was coming for programming jobs, I didn't think they meant creating stupid, conflict-driven drama too: github.com/matplotlib/m...
Seriously, launch all this AI-driven bullshit into the sun. Or give us a way to hold those behind these agents and their generated idiocy accountable.
We're continually updating our website. Here are recent highlights
Sigstore attestations and provenance information are much more visible.
This now makes sense to me! 😭
I just made a sidecar viewer of output you can run in jupyter console or even a regular ipython terminal to get rich interactives without a notebook.
pip install runtimed
import runtimed
runtimed.sidecar()
The @uwssec.bsky.social team wrote up why they switched to Pixi for all their research software engineering: "Pixi has allowed us to remove the friction of package management to focus on what matters: writing better, reproducible scientific software."
prefix.dev/blog/seeking...
The new engine in Python-Blosc2 4.0 delivers quite awesome performance results. Processing a 400 MB array can be up to 9.6x faster than NumPy 🤯
The best part is that it fully supports NumPy's array and ufunc interfaces. High performance with zero friction! 🏎️💨
ironarray.io/blog/miniexp...
Enjoy!
Clean desktop showing Pixi GUI app window listing build tasks and a terminal window displaying compilation logs on a bright yellow background
We're excited to announce Pixi GUI, developed by our working student Felix Häcker! With the goal to make Pixi accessible to a wider audience while also empowering expert users.
Find out more in Felix's blog post: prefix.dev/blog/introd...
Written by an AI?
🧪 Awesome announcement from the @quantstack.bsky.social team! Introducing notebook.link, share your computational notebook and have them run directly in the browser with no setup, powered by Jupyterlite and the xeus stack.
#datascience #scientificcomputing
New open source: cuthbert 🐛
State space models with all the hotness: (temporally) parallelisable, JAX, Kalman, SMC
The Signal app icon.
When it comes to private communications, I trust Signal over everything else.
Toad-Pythonic Universal Terminal Interface for AI Agents-"
Wouldn't it be great to have an interface that could work with any AI agent? #Toad is #Python 's answer to a unified front-end for the many #AI #coding agents out there"
cutt.ly/ytxPsRR2
@willmcgugan.bsky.social @pythonhub.dev
