Trying to understand the training of PG-GAN on LSUN dataset with class labels

[prash030]

Hi @aleju ,
A big thank you for the summary of the PG-Gan paper. I am trying to understand this method and your repository was really helpful!

I have a question, and I was wondering if I can get your thoughts: The paper mentions that their training is unsupervised - meaning that it was not label-conditioned. Then how come they were able to generate label-specific images for LSUN dataset? Did they train separate networks for each label or is their network a multi-class generator?

Thanks in advance!

[aleju]

I assume they trained one network per label, which is kind of the standard for LSUN. Usually you would only get examples from a model trained on the category "bedroom" and/or one trained on "churches", but they seem to have trained one model for each category.

[prash030]

Thank you!