You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Training models should be selected based on evaluation loss on the validation set. As for the eval metric, it should be applied and displayed for users' further information as well.
2. Motivation
The criterion for selecting the best model could be the higher the better, or the lower the better, making things complex. To unify and simplify this, just apply loss on the validation set for model selection. If eval metric is provided, calculate the value and display it for users as well.
3. Your contribution
Will make a PR
The text was updated successfully, but these errors were encountered:
1. Feature description
Training models should be selected based on evaluation loss on the validation set. As for the eval metric, it should be applied and displayed for users' further information as well.
2. Motivation
The criterion for selecting the best model could be the higher the better, or the lower the better, making things complex. To unify and simplify this, just apply loss on the validation set for model selection. If eval metric is provided, calculate the value and display it for users as well.
3. Your contribution
Will make a PR
The text was updated successfully, but these errors were encountered: