- Yes, the weighted loss can be reused to get a signal about overfitting as @KFrank explains in this post.
- Both datasets should come from the same domain and have the same or similar distributions as also explained in the linked post.
- For the final run on the test data you should calculate the metric which represents your target the closest. E.g. calculating the accuracy for an imbalanced use case could show a high value, while your model might be useless and predict only the majority class due to the accuracy paradox.
1 Like