3-32，3-33Pytorch与autograd中的几个重要概念

只有叶子张量才有梯度

import torch
class line(torch.autograd.Function):
    @staticmethod
    def forward(ctx,w,x,b):
        #y = w*x + b
        ctx.save_for_backward(w,x,b)
        return w * x + b
    @staticmethod
    def backward(ctx,grad_out):
        w,x,b = ctx.saved_tensors

        grad_w = grad_out * x
        grad_x = grad_out * w
        grad_b = grad_out
        return grad_w,grad_x,grad_b

w = torch.rand(2,2,requires_grad=True)
x = torch.rand(2,2,requires_grad=True)
b = torch.rand(2,2,requires_grad=True)

out = line.apply(w,x,b)
out.backward(torch.ones(2,2))
print(w,x,b)
print(w.grad,x.grad,b.grad)

tensor([[0.9691, 0.7199],
        [0.3945, 0.3675]], requires_grad=True) tensor([[0.3341, 0.2925],
        [0.3486, 0.7383]], requires_grad=True) tensor([[0.5318, 0.2300],
        [0.7061, 0.9636]], requires_grad=True)
tensor([[0.3341, 0.2925],
        [0.3486, 0.7383]]) tensor([[0.9691, 0.7199],
        [0.3945, 0.3675]]) tensor([[1., 1.],
        [1., 1.]])