Popular Courses
Linear Regression
Introduction
Decision Trees
Introduction
Neural Networks
Introduction

Sign in to Curvingo

or

New Account     Reset Password     

Testing Your AI
Introduction
Our Previous Approach
K-Fold Cross-Validation
Leave-One-Out Cross Validation (LOOCV)
See For Yourself
Which K Should You Pick?
Conclusion
Built with Mathigon

Share

Send us Feedback

Please let us know if you have any feedback and suggestions, or if you find any errors and bugs in our content.

Sorry, your message couldn’t be submitted. Please try again!

Thanks for your feedback!

Reset Progress

This will delete your progress and chat data for all chapters in this course, and cannot be undone!

Glossary

Select one of the keywords on the left…

Testing Your AISee For Yourself

Reading time: ~5 min

To make K-Fold Cross-Validation crystal clear, go ahead and test it yourself! In the interactive simulation below, we are using a simple linear regression model.

Try grabbing the slider and dragging the value of k to change the number of folds used in the Cross-Validation. Watch what happens as the original dataset gets sliced and rotated:

Because each "fold" is forced to study a slightly different combination of data points, every single fold produces a slightly different trendline. The magical final test grade is completely evaluated by taking the performance across all the folds in aggregate!

A Hidden Tradeoff!

While playing around with the simulation, you may have observed something interesting: the lines of best fit wildly jump around far more for purely low values of k than they do for higher values.

This strange behavior isn't an optical illusion. It is actually a direct result of our old archnemesis: The Bias-Variance Tradeoff. Let's break down exactly why this happens in the next section!

Sina