Questions about using multi gpus

[Jiahang Wang]

Hi, I am confused with the code below:

n_gpus = opt.n_gpus_gen if opt.batchSize == 1 else 1 # number of gpus used for generator for each batch

1. I don't know why n_gpus should be 1 when batchsize > 1.
2. the meaning of putting the generator and discriminator on different gpus while training, instead of directly distribute the gens and dis on the multi gpus.
   self.split_gpus = (self.opt.n_gpus_gen < len(self.opt.gpu_ids)) and (self.opt.batchSize == 1)

Need your help! Thanks in advance.