02-pytorch

import torch
from pprint import pprint
from torch.autograd import Variable

变成图纸中的一个节点

tensor = torch.FloatTensor([[1,2],[3,4]])
variable = Variable(tensor,requires_grad=True)
pprint(tensor)
pprint(variable)

tensor([[1., 2.],
        [3., 4.]])
tensor([[1., 2.],
        [3., 4.]], requires_grad=True)

反向传播误差

t_out = torch.mean(tensor*tensor)  # 求x^2的平均值
v_out = torch.mean(variable*variable)
# v_out = 1/4 *sum(var*var) 
v_out.backward()  # 反向传播误差
# d(v_out)/d(var)    1/4*2*variable = variable/2

print(variable.grad)

tensor([[0.5000, 1.0000],
        [1.5000, 2.0000]])

# variable.data 才是 tensor 的形式
print(variable.data)

tensor([[1., 2.],
        [3., 4.]])