model.to(device) 和 model=model.to(device) 有什么区别? [英] What is the difference between model.to(device) and model=model.to(device)?

问题描述

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

Suppose the model is originally stored on CPU, and then I want to move it to GPU0, then I can do:

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

这两行有什么区别?

推荐答案

没有语义差异.nn.Module.to 函数将模型移动到设备.

No semantic difference. nn.Module.to function moves the model to the device.

但要小心.

对于张量(文档):

# 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(文档):

# 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 

这篇关于model.to(device) 和 model=model.to(device) 有什么区别?的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！