Lesson 21 — Logistic prediction

Logistic regression: score, probability, class

Logistic regression is the classic yes/no model. Step 1: a score z = dot product of weights and features, plus a bias. Step 2: sigmoid(z) turns the score into a probability. Step 3: compare with a threshold (0.5) to choose the class.

score = sum(c * v for c, v in zip(coefs, values)) + offset

Each weight says how much a feature pushes towards "yes" (positive) or "no" (negative).

Why it matters: this three-step recipe is exactly what the last layer of most neural networks does for classification.