WRITE · TRACE · UNDERSTAND
Logistic Regression
Turn a score into a probability and learn a binary boundary
Your experiment
fit(xs, ys, steps, rate) starts w=b=0 and returns (w, b). Use sigmoid(w*x+b) and gradients mean((p-y)*x), mean(p-y) for binary cross-entropy. Inputs in this lab have moderate scores.
Python runs in your browser. Your draft stays on this device.
What to look for
The sigmoid maps a real score into (0,1). For binary cross-entropy its derivative with respect to the score is p-y. The trace shows the parameters and mean squared probability residual as a diagnostic, not as the optimized loss.
Watch your code work
Ready to experimentWrite a solution or load the worked answer, then choose Trace & check.
Step through arrays, variables and training curves produced by your own Python.