The CGAN class contains all the functions necessary for running a conditional GAN based on the CGAN model. The deep convolutional generative adversarial networks proved to have the performance in generating photo-like quality outputs. We have previously introduced CGANs, so just to remind you, their reference paper is:
RADFORD, Alec; METZ, Luke; CHINTALA, Soumith. Unsupervised representation learning with deep convolutional Generative Adversarial Networks. arXiv preprint arXiv:1511.06434, 2015 at https://arxiv.org/abs/1511.06434.
In our project, we will then add the conditional form of the CGAN that uses label information as in a supervised learning task. Using labels and integrating them with images (this is the trick) will result ...
