Broad intuition · Study map 9 of 85
Linear models, trees, neighbors, and ensembles
Suggested preparation: Probability and loss: how surprising was the answer?
Picture it
Different models carve up the input space differently: a line separates regions, a tree asks a sequence of questions, and neighbors vote from nearby examples. Ensembles combine many imperfect views.
What to understand
- Linear/logistic regression, k-nearest neighbors, SVMs, and kernels
- Decision trees, random forests, gradient boosting, and calibration
Follow the references
1 · Documentation
scikit-learn · Supervised learning ↗Compare linear models, nearest neighbors, SVMs, and ensembles using the examples and diagrams.
You understand it when…
For a small table of customer features, what would you try before a large neural network, and how would you choose?
This is a study map. Follow the reference for the full explanation; use the question to test your mental model. Browse the reference shelf.