Click the scatter plot to add points (toggle class with dropdown). Train a simple logistic model, then inspect ROC/AUC and PR curves. AUC summarizes ranking quality independent of a fixed threshold.
Every coffee helps keep the servers running. Every book sale funds the next tool I'm dreaming up. You're not just supporting a site — you're helping me build what developers actually need.
This module computes ROC curves by sweeping probability thresholds over predicted scores and true labels. It calculates TPR/FPR pairs, area under the curve (AUC) via trapezoidal rule, and can display class‑imbalance aware views (e.g., PR curve where available).