pytorch中.to(device) 和.cuda()的区别说明

device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu") # 单GPU或者CPU
model.to(device)
#如果是多GPU
if torch.cuda.device_count() > 1:
  model = nn.DataParallel(model，device_ids=[0,1,2])
model.to(device)

#指定某个GPU
os.environ['CUDA_VISIBLE_DEVICE']='1'
model.cuda()
#如果是多GPU
os.environment['CUDA_VISIBLE_DEVICES'] = '0,1,2,3'
device_ids = [0,1,2,3]
net  = torch.nn.Dataparallel(net, device_ids =device_ids)
net  = torch.nn.Dataparallel(net) # 默认使用所有的device_ids 
net = net.cuda()

class DataParallel(Module):
    def __init__(self, module, device_ids=None, output_device=None, dim=0):
        super(DataParallel, self).__init__()

        if not torch.cuda.is_available():
            self.module = module
            self.device_ids = []
            return

        if device_ids is None:
            device_ids = list(range(torch.cuda.device_count()))
        if output_device is None:
            output_device = device_ids[0]

# 开始脚本，创建一个张量
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
...
# 但是无论你获得一个新的Tensor或者Module
# 如果他们已经在目标设备上则不会执行复制操作
input = data.to(device)
model = MyModule(...).to(device)