with torch.no_grad()

在pytorch中,tensor有一个requires_grad参数,如果设置为True,则反向传播时,该tensor就会自动求导。tensor的requires_grad的属性默认为False,若一个节点(叶子变量:自己创建的tensor)requires_grad被设置为True,那么所有依赖它的节点requires_grad都为True(即使其他相依赖的tensor的requires_grad = False)

 

x = torch.randn(10, 5, requires_grad = True)
y = torch.randn(10, 5, requires_grad = False)
z = torch.randn(10, 5, requires_grad = False)
w = x + y + z
print(w.requires_grad)

True

with torch.no_grad
上文提到volatile已经被废弃,替代其功能的就是with torch.no_grad。作用与volatile相似,即使一个tensor(命名为x)的requires_grad = True,由x得到的新tensor(命名为w-标量)requires_grad也为False,且grad_fn也为None,即不会对w求导。例子如下所示:

 
x = torch.randn(10, 5, requires_grad = True)
y = torch.randn(10, 5, requires_grad = True)
z = torch.randn(10, 5, requires_grad = True)
with torch.no_grad():
    w = x + y + z
    print(w.requires_grad)
    print(w.grad_fn)
print(w.requires_grad)


False
None
False

 

posted @ 2022-03-09 10:52  Tomorrow1126  阅读(70)  评论(0编辑  收藏  举报