Implementation of Dual Wasserstein GAN for face2anime.

(https://arxiv.org/pdf/1704.02510.pdf)

(https://arxiv.org/pdf/1704.00028.pdf)

Works with 64 by 64 pictures of I don't remember which datasets that I found on google.

TODO: make train.py be device-agnostic.

TODO: log losses and make some better visualizations.

UPD (june 15): managed to get gtx 1080 for a few days and currently it is training.