Hancheng Ye

Hi, thanks for ur attention to our work. It seems that self.model.cuda().info(False, self.inputsize) return a Nonetype object. But from [model_info()](https://github.com/HankYe/PAGCP/blob/4a81b863a5ec4d95f39d695ccc43bc982b7604de/utils/torch_utils.py#L243) function, it should return a tuple. So I suggest you...

Hello, thanks for ur attention to our work. To resume training the pruned model, you can add some codes to load the optimizer states of the pruned model after [line...

Hello, thanks for ur attention to our work. It seems to be the communication issue between different GPUs in the cluster, which might be caused by network latency or load...