我不理解代码"because weights have requires_grad=True, but we don't need to track this in autograd"中第60-61行中的注释我知道我们提到的变量是requires_grad=True,我们需要这些变量来计算使用自梯度的梯度,但是成为"tracked by autograd"意味着什么呢?
https://colab.research.google.com/github/pytorch/tutorials/blob/gh-pages/_下载/神经_网络_tutorial.ipynbloss = criterion(output, target)optimizer.step() # Does the update
从about代码<e