Interactive ROC, AUC & Precision–Recall

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.


Dataset & Model
Class 1 Class 0 Decision boundary w₀x + w₁y + b = 0
ROC Curve
Each point = threshold. Diagonal = random (AUC 0.5)
Precision–Recall Curve
PR is more informative on imbalanced data
Controls
Metrics
Log loss
Accuracy
ROC AUC
PR AUC
N points
0

Support This Free Tool

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.

500K+ users
200+ tools
100% private
Privacy Guarantee: Private keys you enter or generate are never stored on our servers. All tools are served over HTTPS.

About This Tool & Methodology

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).

Learning Outcomes

  • Interpret ROC curves and AUC across different thresholds.
  • Understand when ROC vs PR curves are appropriate (imbalance).
  • See how calibration and score distributions affect metrics.

Authorship & Review

  • Author: 8gwifi.org engineering team
  • Reviewed by: Anish Nath
  • Last updated: 2025-11-19

Trust & Privacy

  • All calculations happen in your browser; datasets are not uploaded.