Out-of-Sample Testing for GANs
2019-01-28Code Available0· sign in to hype
Pablo Sánchez-Martín, Pablo M. Olmos, Fernando Pérez-Cruz
Code Available — Be the first to reproduce this paper.
ReproduceCode
- github.com/psanch21/EvalGANOfficialIn papertf★ 3
- github.com/yangyucheng000/Paper-2/tree/main/Out-of-sample-restoration-mainmindspore★ 0
- github.com/yangyucheng000/papercode-2/tree/main/Out-of-sample-restoration-mainmindspore★ 0
Abstract
We propose a new method to evaluate GANs, namely EvalGAN. EvalGAN relies on a test set to directly measure the reconstruction quality in the original sample space (no auxiliary networks are necessary), and it also computes the (log)likelihood for the reconstructed samples in the test set. Further, EvalGAN is agnostic to the GAN algorithm and the dataset. We decided to test it on three state-of-the-art GANs over the well-known CIFAR-10 and CelebA datasets.