Cross-Validation Learning Tool

Learn how cross-validation works by fitting a model to different subsets of data.

Step 1: Choose Your Setup

We've generated 36 data points from a quadratic function with some noise. How many ways do you want to split the data for cross-validation?

Fold 1 of N

Use the sliders to fit a curve (`y = dx³ + ax² + bx + c`) to the blue training data. Minimize the training error.

Training RMSD:
0.00
Test RMSD:
0.00

Cross-Validation Summary

Fold Cubic (d) Quadratic (a) Linear (b) Intercept (c) Training RMSD Test RMSD

Average Test RMSD:

This average value gives you a more robust estimate of your model's performance on unseen data.