Overfitting

Overfitting is the biggest issue for training a model, and occurs when a model is trained on its data so much that it no longer generalizes well. This is far mor dangerous than underfitting, as you may actually deploy an overfitted model if you arent careful (or your coworkers might!). Fixes for this are many, and include using techniques such as ridge regression and feature selection.

Last Updated: April 07, 2019