comparison against stylegan2-ada

mit-han-lab / data-efficient-gans

[NeurIPS 2020] Differentiable Augmentation for Data-Efficient GAN Training

https://arxiv.org/abs/2006.10738

BSD 2-Clause "Simplified" License

1.27k stars 175 forks source link

Closed molo32 closed 3 years ago

molo32 commented 3 years ago

how does the precision of this job compare vs stylegan2-ada.

zsyzzsoft commented 3 years ago

We included such a comparison in Table 3 in the newest version of our paper.