Distributed Adversarial Examples Generation

[wanliAlex]

Hi,

The adversarial examples generation is always time-consuming in adversarial training. Is there a way to conduct multi-GPU generation in tensorflow2?

e.g. For a batch with 128 samples using 4 GPUs, can I generate 128/4 adversarial examples by each GPU at the same time and concatenate them later?

Look forward to hearing from you.