Lightnews — Scholar-powered news

Riccardo Cappuzzo @riccardocappuzzo.com · 1d

"ok the test run is done, let's see"

...

"this will be hard to debug"

Reposted by Riccardo Cappuzzo

Emilien Schultz @emilienschultz.bsky.social · 10d

What a banger is skrub @skrub-data.bsky.social !

Big thumbs up for the sklearn team & the maintainer of this package

1 4 12

Riccardo Cappuzzo @riccardocappuzzo.com · 3d

Thanks a lot for the compliments! I had a lot of fun giving the talk, and I'm happy to see people liked it

1 2

Riccardo Cappuzzo @riccardocappuzzo.com · 15d

My first actual talk in front of a ton of people 🙃

Skrub @skrub-data.bsky.social · 15d

📅 Less than a week away! The talk will be on Oct 1st at 10.05AM in room Louis Armand 1 - Est.

If you want to contribute to skrub, we will also have a sprint on Thursday.

See you there!

PyData Paris @pydataparis.bsky.social · Aug 12

📢 Talk Announcement

"Skrub: machine learning for dataframes", by Guillaume Lemaitre, Jérôme Dockès and @riccardocappuzzo.com.
@skrub-data.bsky.social

📜 Talk info: pretalx.com/pydata-paris-2025/talk/T9KTPU
📅 Schedule: pydata.org/paris2025/schedule
🎟 Tickets: pydata.org/paris2025/tickets

Riccardo Cappuzzo @riccardocappuzzo.com · Sep 5

TIL about lava lamp encryption
www.cloudflare.com/learning/ssl...

How do lava lamps help with Internet encryption?

The Cloudflare lava lamps are used for Internet encryption. Learn about entropy in cryptography and why randomness is essential for SSL encryption.

www.cloudflare.com

1

Reposted by Riccardo Cappuzzo

Skrub @skrub-data.bsky.social · Sep 5

Do you have to deal with numerical features that involve large outliers, and need to train linear models or neural networks?

Then you might want to try the skrub SquashingScaler. The SquashingScaler behaves like scikit-learn RobustScaler, but smoothly clips outliers to predefined boundaries.

1 1 1

Riccardo Cappuzzo @riccardocappuzzo.com · Sep 4

context: www.youtube.com/watch?v=f7Mi...

The Powerpuff Girls | Theme Song | Cartoon Network

YouTube video by The Powerpuff Girls

www.youtube.com

Riccardo Cappuzzo @riccardocappuzzo.com · Sep 4

Working hard on the next @skrub-data.bsky.social slide deck...

1 3

Reposted by Riccardo Cappuzzo

Olivier Grisel @ogrisel.bsky.social · Aug 19

Today at #EuroScipy2025, @glemaitre58.bsky.social and I presented a tutorial on pitfalls of machine learning for imbalanced classification problems.

We discussed what (not) to do when fitting a classifier and obtaining degenerate precision or recall values.

probabl-ai.github.io/calibration-...

Imbalanced classification: pitfalls and solutions — Probabilistic calibration of cost-sensitive learning

probabl-ai.github.io

1 10 23

Reposted by Riccardo Cappuzzo

PyData Paris @pydataparis.bsky.social · Aug 12

📢 Talk Announcement

"Skrub: machine learning for dataframes", by Guillaume Lemaitre, Jérôme Dockès and @riccardocappuzzo.com.
@skrub-data.bsky.social

📜 Talk info: pretalx.com/pydata-paris-2025/talk/T9KTPU
📅 Schedule: pydata.org/paris2025/schedule
🎟 Tickets: pydata.org/paris2025/tickets

1 4

Reposted by Riccardo Cappuzzo

Olivier Grisel @ogrisel.bsky.social · Aug 18

Attending the @skrub-data.bsky.social tutorial by @riccardocappuzzo.com and @glemaitre58.bsky.social at #EuroScipy2025. They introduce the new DataOps feature released in skrub 0.6.

Here is the repo with the material for the tutorial: github.com/skrub-data/E...

Photo of Riccardo presenting skrub DataOps in a lecture room to an audience of ~50 people.

2 5

Reposted by Riccardo Cappuzzo

Mike Fiedler @miketheman.com · Jul 28

Heads Up, #Python Developers!

There is an active phishing attack targeting PyPI users.

• Threat: Emails from [email protected] (with a 'j') link to a fake login page.
• Action: Do not click any links. If you already did, change your PyPI password ASAP.
• Note: PyPI itself has not been breached.

2 40 59

Riccardo Cappuzzo @riccardocappuzzo.com · Jul 24

"My watch gained gemini power overnight"

One sentence horror

Riccardo Cappuzzo @riccardocappuzzo.com · Jul 24

Huge release, and the first one where I felt like I actually contributed a lot to the final result.

I really think DataOps are a game changer, and I can't wait to see what people come up with with them.

I also ended up rewriting most of the user guide, hopefully improving it along on the way 😂

Skrub @skrub-data.bsky.social · Jul 24

⚡ Release 0.6.0 is now out! ⚡

🚀 Major update! Skrub DataOps, various improvements for the TableReport, new tools for applying transformers to the columns, and a new robust transformer for numerical features are only some of the features included in this release.

3 4

Riccardo Cappuzzo @riccardocappuzzo.com · Jul 19

The walrus operator mystifies me

1

Riccardo Cappuzzo @riccardocappuzzo.com · Jul 19

I got 16/26 on the fstrings.wtf quiz. Can you do better? fstrings.wtf

fstrings.wtf - Python F-String Quiz

Test your knowledge of Python's f-string formatting with this interactive quiz. How well do you know Python's string formatting quirks?

fstrings.wtf

1

Reposted by Riccardo Cappuzzo

Randall Munroe @xkcd.com · Jun 19

David R. Hagen just solved a small mystery that I mentioned 13 years ago in the mouseover text of a comic drhagen.com/blog/the-mis...

The Missing 11th of the Month - David R Hagen

Personal website of David R Hagen, scientific software engineer

drhagen.com

60 610 3.2K

Riccardo Cappuzzo @riccardocappuzzo.com · Jun 13

Really cool graffiti I spotted while walking around in the town where I live

1

Riccardo Cappuzzo @riccardocappuzzo.com · Jun 13

classic

arstechnica.com/ai/2025/06/g...

AI Overviews hallucinates that Airbus, not Boeing, involved in fatal Air India crash

Google’s disclaimer says AI “may include mistakes,” which is an understatement.

arstechnica.com

Riccardo Cappuzzo @riccardocappuzzo.com · Jun 4

Out of all the features in the expressions, this may be my personal favorite. I always end up adding too many configurations just because the syntax is so convenient.

Skrub @skrub-data.bsky.social · Jun 4

👀 This week's post will be another sneak peek into skrub expressions, an upcoming feature that will ease the preparation and execution of machine learning pipelines on dataframes.

This time we will focus on how expressions can simplify the construction of complex hyperparameter grids.

Reposted by Riccardo Cappuzzo

Skrub @skrub-data.bsky.social · May 28

📝 The skrub TextEncoder brings the power of HuggingFace language models to embed text features in tabular machine learning, for all those use cases that involve text-based columns.

2 3 6

Riccardo Cappuzzo @riccardocappuzzo.com · May 23

www.justfuckingcode.com

Just fucking code.

www.justfuckingcode.com

Reposted by Riccardo Cappuzzo

Skrub @skrub-data.bsky.social · May 21

The Skrub Cleaner is a lightweight transformer that performs consistency checks on a dataframe:

🔍 It gives a uniform representation of null values, converting those represented as strings (such as "N/A")
🗑️ It drops columns that contain too many null values (according to a user-defined threshold)

1 2 6

Riccardo Cappuzzo @riccardocappuzzo.com · May 20

Now that the paper is out, I can finally share the totally-not-confusing script/plot/table map I made to track which scripts prepare which figures and tables and from what data.

If it wasn't clear, don't do this. If you *really* have to, I used the @obsidian.md canvas for this.

5

Riccardo Cappuzzo @riccardocappuzzo.com · May 20

Tech issues happen 🙈 this is what it should look like bsky.app/profile/ricc...

And thanks!

Riccardo Cappuzzo @riccardocappuzzo.com · May 19

A bit of a mess up with this figure! This is what it's supposed to look like 🙈

1