In supplementary materials, I found that what types of GPUs are used to obtain the results, but any information about training times could not be found.
Can you provide approximate times (or days) to train the model for ImageNet 64/128/256 for both unconditional and conditional BigGANs and COCO-stuff 128, 256 for StyleGAN2?
Thank you for sharing the code of this great work!