Common preprocessing steps

<aside> 💡 A model is only as good as its features.

</aside>

<aside> 💡 Use Polars for feature engineering. At least for the heavy stuff.

</aside>

Ideen für Feature Engineering (checklist)

• Feature-Engineering-Checklist.pdf
• Checkliste
• Fastai Buch S. 559
• The most important techniques for feature engineering (The Kaggle Book)
• Unsupervised clustering to cluster observations into groups. The end result is a categorical feature with similar observations being members of the same group.
• Easily derived features

Feature Engineering

Feature-Engine

Dimensionality Reduction

Dimensionality Reduction

Working with datetime

Time Feature Engineering