Decision trees and forests

Decision trees split data with if/then rules. Random forests average many trees trained on random slices—strong tabular baselines.

Readable baselines and strong ensembles before jumping to deep nets on spreadsheets.

Pick splits that purify labels → grow until stop rules → forests bag trees and votes. Watch overfitting on deep single trees.

Hand-build a 2-split tree for a tiny yes/no dataset on paper.

⚠️ Myth: Trees are always interpretable in forests.

✓ Reality: Single short trees are; huge forests are opaque without tools.

• Course 03 classical-ml-models; ensembles-boosting-deeper
• scikit-learn: https://scikit-learn.org/ ↗