In Pipeline comparision tab, what is the use of holdout data set

[kbharat439]

Please explain the use of score a holdout data set.

[DannyModlin · Accepted solution]

Greetings.

The use of holdout data is to see how well a model will generalize and perform on data that was not trained upon. This aids in the prevention of overfitting a model and being subjected to a false sense of accuracy in how well a model is performing.

Hope this helps.