Principal component analysis
Fit reusable axes, choose retained variation and interpret weights, scores and the limits of a two-dimensional view.
Helpful prior knowledge: Distance depends on scale and context. These links are guidance, not locks.
Follow teaches the idea; Change asks you to make a choice; Transfer changes the context. Apply, reviews and checkpoints appear where the existing lessons provide them. Go Further lessons are optional.
Lessons in this model deck
Start with the first core lesson →
New coordinates, not selected original columns
Distinguish PCA from feature selection and clustering.
12–18 min · 3 practicesFit a reusable PCA representation
Transform rows with the fitted axes and scale.
20–30 min · 4 practicesExplained variance is not predictive accuracy
Interpret per-component variation.
12–18 min · 3 practicesSelect a retained representation
Choose the smallest component prefix meeting a variance criterion.
12–18 min · 3 practicesComponent weights and scores
Distinguish feature-axis weights from row coordinates.
12–18 min · 3 practicesTwo dimensions are a view
Separate a display from the retained representation.
12–18 min · 3 practicesPCA retrieval
Retrieve earlier concepts before combining them.
15–20 min · 3 practicesEquivalent signs and reconstruction
Understand equivalent representations and information loss.
18–25 min · 4 practicesPCA checkpoint
Fit Penguin PCA, retain at least 90% variance, label weights and report the separate 2D view.
25–35 min · 1 practices