How do you deal with catastrophic forgetting?

Working with transfer learning or online learning, how do you ensure that the model is improving with each iteration and not forgetting important things

Hi @guandaline! :wave:

One way to check this is to make sure that all of the classes of data that you care about (including the “old” data) are represented in your test set and test on it.

To ensure that the model won’t forget important things, you should include “old” samples that represent cases that you care about in your “new” training set.