原理

.to(device) 可以指定CPU 或者GPU

device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu") # 單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)

.cuda() 只能指定GPU

#指定某個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) # 默認使用所有的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 = moduleself.device_ids = []returnif device_ids is None:device_ids = list(range(torch.cuda.device_count()))if output_device is None:output_device = device_ids[0]