Lightnews — Scholar-powered news

skrub-data.org/stable/refer...

Skrub @skrub-data.bsky.social · 20h

ApplyToCols

Gallery examples: Getting Started Hands-On with Column Selection and Transformers

Hands-On with Column Selection and Transformers

1 1

Skrub @skrub-data.bsky.social · 20h

Example: skrub-data.org/stable/auto_...

In previous examples, we saw how skrub provides powerful abstractions like TableVectorizer and tabular_pipeline() to create pipelines. In this new example, we show how to create more flexible pipel...

Skrub @skrub-data.bsky.social · 20h

For even more control over column selection, skrub provides a collection of selectors that let you partition dataframes by data type, column name, or user-specified functions.

Skrub @skrub-data.bsky.social · 20h

All these transformers can be concatenated and inserted in a scikit-learn pipeline to build a feature matrix with complex column selection operation, and can be seen as an alternative for the scikit-learn ColumnTransformer.

Skrub @skrub-data.bsky.social · 20h

ApplyToFrame selects columns in the same way, but then uses all of them at the same time as input to the transformer: this is useful for dimensionality reduction.
SelectCols and DropCols can be used as "filtering blocks" in a pipeline.

Skrub learning materials – Skrub

Skrub @skrub-data.bsky.social · 20h

Skrub includes a powerful set of transformers and selectors that allow to transform columns based on various conditions.

ApplyToCols lets you select a subset of columns in your dataframe, then applies a transformer to each selected column separately.

1 2

Reposted by Skrub

Léo Mignot @leomignot.bsky.social · 1d

On vous a déjà dit que Skrub c'est cool ? Et que l'intervention de @riccardocappuzzo.com était très chouette ? Hein, on vous l'a dit ?
skrub-data.org/skrub-materi...

1 3

Skrub @skrub-data.bsky.social · 1d

Slides:
skrub-data.org/skrub-materi...

Skrub @skrub-data.bsky.social · 1d

Thanks to @riccardocappuzzo.com , @glemaitre58.bsky.social and Jérôme Dockès for preparing the talk, and mentoring at the sprint!

Skrub @skrub-data.bsky.social · 1d

The sprint was also a big hit, with both new and old contributors working on issues and getting to know the repository.

And to cap it all off, thanks to P16 we have stickers now 🚀

Skrub @skrub-data.bsky.social · 1d

@pydataparis.bsky.social 2025 is over, and it was a big success!

Our talk was very well received, and we got a lot of great questions, especially about scalability and how to interface with other libraries in production environments.

The skrub sticker on the back of a laptop

1 3

Reposted by Skrub

Emilien Schultz @emilienschultz.bsky.social · 8d

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

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

1 4 12

Skrub @skrub-data.bsky.social · 13d

📅 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

1 4

Skrub @skrub-data.bsky.social · 13d

🛠️ Main bugfixes
- Fixed the display of DataOp objects in Google Colab cell outputs.
- Fixed the range from which choose_float and choose_int sample values when log=False and n_steps is None.
- The SkrubLearner used to do a prediction on the train set during fit(), this has been fixed.

1 1

Skrub @skrub-data.bsky.social · 13d

👀 Changes and deprecations
- Ken embeddings are now deprecated.
- The accepted values for the parameter how of .skb.apply() have changed. The new values are "auto", "cols", "frame", and "no_wrap".
- The parameter splitter of .skb.train_test_split() has been renamed split_func.

Skrub @skrub-data.bsky.social · 13d

🚀 New features
- The DataOp.skb.full_report() now displays the time each node took to evaluate.
- The User guide has been reworked and expanded.

Release 0.6.2 · skrub-data/skrub

Skrub @skrub-data.bsky.social · 13d

⚡ Release 0.6.2 is out ⚡

github.com/skrub-data/s...

New features The DataOp.skb.full_report() now displays the time each node took to evaluate. #1596 by Jérôme Dockès. The User guide has been reworked and expanded. Changes and deprecations Ken em...

github.com

1 4 5

Reposted by Skrub

Léo Mignot @leomignot.bsky.social · 26d

Reminder: skrub == cool

Skrub @skrub-data.bsky.social · 26d

skrub DataOps help you construct complex and extensive hyperparameter search spaces. However, interpreting results from large grids can be challenging.
To address this, skrub generates a parallel coordinate plot that visualizes all runs and the parameters used to achieve specific results.

4 8

Skrub @skrub-data.bsky.social · 26d

Here's another example on how to tune ML models with skrub Data Ops: skrub-data.org/stable/auto_...

Hyperparameter tuning with DataOps

A machine-learning pipeline typically contains some values or choices which may influence its prediction performance, such as hyperparameters (e.g. the regularization parameter alpha of a RidgeClas...

Skrub DataOps applied to forecasting timeseries — Skrub DataOps applied to forecasting timeseries

1 1

Skrub @skrub-data.bsky.social · 26d

The plot in the video was created for our EurosciPy 2025 tutorial on forecasting time series: skrub-data.org/EuroSciPy202...

Skrub @skrub-data.bsky.social · 26d

The plot is interactive: you can select a range of results, and it will highlight only the runs within that range, enabling you to refine your search further. It also tracks fit and score times, so you can identify which parameters most impact runtime.

skrub-data.org/stable/refer...

Skrub @skrub-data.bsky.social · 26d

skrub DataOps help you construct complex and extensive hyperparameter search spaces. However, interpreting results from large grids can be challenging.
To address this, skrub generates a parallel coordinate plot that visualizes all runs and the parameters used to achieve specific results.

1 6

Skrub @skrub-data.bsky.social · Sep 5

SquashingScaler

Gallery examples: SquashingScaler: Robust numerical preprocessing for neural networks

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.