SquishyQuokka
27moby
When do you stop training a Neural Network?
- Early cut off looking at training_loss
- Early cut off looking at validation_loss
- Set Higher Regularization?
Increasing data is not an option, have used SMOTE related technqiues for interpolating more data for the imbalanced classes already.
Just trying to understand what do other folks do.
27mo ago
GoofyDonut
27mo
When do you stop training a Neural Network
When it gains sentience
GroovyBoba
Before you max out your credit card due to compute bill.
SquishyQuokka
27mo
This is an old project. Was looking through Tensorboard log for a run I did in 2021, during the pandemic.
GroovyBoba
Ideally you should look at validation loss as that is the true unseen data and would be closest to new unseen real world data.
Discover more
Curated from across
