Evaluating Model Performance with training and test set

Hi, I’m a bit confused.

In this mission: https://app.dataquest.io/m/235/the-linear-regression-model/6/making-predictions I don’t understand why we also calculated the RMSE value with training set and I see that you do the same in the following missions. What is the reason?

Thanks for reading.

1 Like

Hi @arredocana

This is done to detect overfitting.

A model that is underfit will have high training and high testing error while an overfit model will have extremely low training error but a high testing error.
https://towardsdatascience.com/overfitting-vs-underfitting-a-complete-example-d05dd7e19765

Best,
Sahil