Logistic scoring begins with class labels rather than a continuous target. The lesson shows how exact arithmetic prepares a score before any named boundary appears.
highlighted = computed this step
Classes instead of a continuous target
The labels are class 0 or class 1. A score will be converted into one of those class decisions.
y∈{0,1}
Why the setup changes
Regression used a line as a fitted value. Logistic scoring uses a line as a score, then a threshold turns that score into a class decision.
z=wx+b
Summary
This lesson changes the target type but keeps the arithmetic exact. The shown table is still the trust surface.