python - model.to(device) 和 model=model.to(device) 有什么区别？

Question

假设模型原本是存储在CPU上的，然后我想把它移到GPU0上，那么我可以这样做:

device = torch.device('cuda:0')
model = model.to(device)
# or
model.to(device)

这两行有什么区别？

Answer 1

没有语义差异。 nn.Module.to函数将模型移动到设备。
但要谨慎。
对于张量 ( documentation ):

# tensor a is in CPU
device = torch.device('cuda:0')
b = a.to(device)
# a is still in CPU!
# b is in GPU!
# a and b are different 

对于模型 ( documentation ):

# model a is in CPU
device = torch.device('cuda:0')
b = a.to(device)
# a and b are in GPU
# a and b point to the same model