本文主要是PyTorch中Variable变量的一些用法。
import torch
from torch.autograd import Variable
tensor = torch.FloatTensor([[1, 2], [3, 4]])
# 定义Variable, requires_grad用来指定是否需要计算梯度
variable = Variable(tensor, requires_grad = True)
print tensor
print variable
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
1 2
3 4
[torch.FloatTensor of size 2x2]
Variable containing:
1 2
3 4
[torch.FloatTensor of size 2x2]
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
# 计算x^2的均值
tensor_mean = torch.mean(tensor * tensor)
variable_mean = torch.mean(variable * variable)
print tensor_mean
print variable_mean
- 1
- 2
- 3
- 4
- 5
7.5
Variable containing:
7.5000
[torch.FloatTensor of size 1]
- 1
- 2
- 3
- 4
- 5
# variable进行反向传播
# 梯度计算如下:
# variable_mean = 1/4 * sum(variable * variable)
# d(variable_mean)/d(variable) = 1/4 * 2 * variable = 1/2 * variable
variable_mean.backward()
# 输出variable中的梯度
print variable.grad
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
Variable containing:
0.5000 1.0000
1.5000 2.0000
[torch.FloatTensor of size 2x2]
- 1
- 2
- 3
- 4
- 5
# *表示逐元素点乘,不是矩阵乘法
print tensor * tensor
print variable * variable
- 1
- 2
- 3
1 4
9 16
[torch.FloatTensor of size 2x2]
Variable containing:
1 4
9 16
[torch.FloatTensor of size 2x2]
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
# 输出variable中的data, data是tensor
print variable.data
- 1
- 2
1 2
3 4
[torch.FloatTensor of size 2x2]