Skip to learning content

DATA SCIENCE PYTHON PLAYGROUND

Machine Learning · Learn / Refresh

← Choose a deckQUESTIONS · MODELS · EVIDENCE
ROUTE 3A

Regression

Predict quantities with lines, curves and trees.

01 / Lines and evidence

02 / Several predictors

03 / Curves and trees

R07Core · teaching
Make curved features: x and x², Curved fit, Validate itx and x²Curved fitValidate it

Make curved features

Understand expansion before fitting.

12–18 min · 3 practices
R08Core · teaching
Validate polynomial flexibility: More degrees, Train improves, Validate firstMore degreesTrain improvesValidate first

Validate polynomial flexibility

Choose degree inside the complete pipeline.

18–25 min · 4 practices
R09Core · teaching
Trees predict with leaf averages: Split on X, Leaf average A, Leaf average BSplit on XLeaf average ALeaf average B

Trees predict with leaf averages

Learn recursive splits and regression leaves within this branch.

25–40 min · 4 practices
R10Core · teaching
Control tree complexity: Control depth, Larger leaves, Validate errorControl depthLarger leavesValidate error

Control tree complexity

Use depth and minimum leaf size with validation evidence.

12–18 min · 3 practices
R-R2Core · review
Flexibility retrieval: Line or curve, Tree flexibility, Common foldsLine or curveTree flexibilityCommon folds

Flexibility retrieval

Retrieve earlier concepts before combining them.

15–20 min · 3 practices
R11Go Further · teaching
What fitted explanations cannot establish: Prediction works, Association seen, No causal claimPrediction worksAssociation seenNo causal claim

What fitted explanations cannot establish

Match claim strength to the evidence.

12–18 min · 3 practices
R-K1Core · checkpoint
Regression checkpoint: Candy features, Compare models, Final RMSECandyfeaturesComparemodelsFinal RMSE

Regression checkpoint

Compare mixed-input linear regression and a tree on Candy with common training folds, a reference, diagnosis and final evidence.

25–35 min · 1 practices