scikit-learn
Classical machine learning in Python — I use it and gradient-boosted trees as baselines that foundation models have to beat.
scikit-learn is the standard Python library for classical machine learning: preprocessing, linear models, trees, clustering and model evaluation.
How I use it
In Real-time Data Systems, scikit-learn and gradient-boosted trees (LightGBM) serve as the baselines every fancier model must beat — including a fine-tuned time-series foundation model. I also covered it in the Machine Learning with Python (V2) course of the IBM Generative AI Engineering Professional Certificate certificate (classification, regression, clustering).
Why it matters
A strong, cheap baseline is the fastest way to find out whether a complex model is actually adding value.