How to decide the training epochs or early stop condition? about taming-transformers HOT 4 OPEN

Thanks!
The VQGAN benefits greatly from training it as long as possible (provided the data set is large enough and overfitting is a secondary concern), and tuning in the discriminator rather late. When training on ImageNet, for example, I would recommend 3-5epochs without the adversarial loss (but more is better) and then training for at least another 3-5 epochs with the discriminator turned on (again, more=better).

The stopping condition for the transformer is usually when it starts overfitting in terms of NLL on held-out test data.

from taming-transformers.

May I ask how long it usually takes to train on the ImageNet and how many GPUs are used?

from taming-transformers.

Any updates on training times? Costs?

from taming-transformers.

mark

from taming-transformers.