import导入

import torch#基本的torch函数
import torch.autograd as autograd#自动求导
import torch.nn as nn#神经网络类都在这个里面
import torch.nn.functional as F#几乎所有的激励函数
import torch.optim as optim#优化

  创建Tensors

#create 1D vector
V_data = [1., 2., 3.]
V = torch.Tensor(V_data)#我用的是pyCharm编辑器,输入torch给的提示没有Tensor函数,其实是有的
print(V)
1
 2
 3
[torch.FloatTensor of size 3]
#create 2D vector
M_data = [[1., 2., 3.], [4., 5., 6.]]
M = torch.Tensor(M_data)
print(M)
 1  2  3
 4  5  6
[torch.FloatTensor of size 2x3]
#create 3D vector
T_data = [[[1.,2.], [3.,4.]],
          [[5.,6.], [7.,8.]]]
T = torch.Tensor(T_data)
print(T)
(0 ,.,.) =
  1  2
  3  4

(1 ,.,.) =
  5  6
  7  8
[torch.FloatTensor of size 2x2x2]

  获取Tensor部分值 

#我就觉得这里比TensorFlow好用多了QAQ
print(V[0])
print(M[0])
print(T[0])
1.0

 1
 2
 3
[torch.FloatTensor of size 3]


 1  2
 3  4
[torch.FloatTensor of size 2x2]

  产生随机数据 

x = torch.randn((3,4,5))
print(x)
(0 ,.,.) =
  1.4533  0.0593  0.2027 -1.0107 -0.3175
 -0.1847  0.3021  0.0848 -1.2445 -0.5568
 -0.2796 -0.5961 -0.3000 -0.2782  1.4920
  1.4030  1.0875 -0.5814 -1.2006  0.2690

(1 ,.,.) =
 -0.7093 -0.4939  0.7491  0.9133  0.4221
  1.3949  2.5685 -0.4359 -0.7788  1.0251
  1.6232 -1.2432  0.3403 -1.0551  1.3790
 -1.5632 -0.9772  0.3963 -0.1890  0.0032

(2 ,.,.) =
 -0.3360 -0.5571 -0.6641 -1.5845 -0.8766
 -0.1809 -1.0035  1.7093  0.9176  1.6438
  1.6955  0.6816  0.5978 -0.1379 -0.3877
  1.0876  1.2371 -0.7378 -0.7647  0.0544
[torch.FloatTensor of size 3x4x5]

  Tensor运算 

x = torch.Tensor([1., 2., 3.])
y = torch.Tensor([4., 5., 6.])
z = x + y
print(z)
 5
 7
 9
[torch.FloatTensor of size 3]

  [res] torch.cat( [res,] x_1, x_2, [dimension] ) 

x_1 = torch.randn(2, 5)
y_1 = torch.randn(3, 5)
z_1 =torch.cat([x_1, y_1])#没有最后一个参数,默认是0,则最终维度的第0维度为x_1与y_1第0维度的和,最终维度的其他维度不变.以下同理
print(z_1)

x_2 = torch.randn(2, 3)
y_2 = torch.randn(2, 5)
z_2 = torch.cat([x_2, y_2], 1)
print(z_2)
0.6372  0.7380  0.9324  0.0626 -0.3678
 1.1819  2.1591  0.2445  0.0064  0.7760
-0.7765 -0.6797  0.1814  0.3948  1.7398
-0.2957 -0.6972  3.7052 -0.1943  0.4159
 0.7385 -0.2365  1.4243 -0.0044 -0.7645
[torch.FloatTensor of size 5x5]


-0.0256 -0.6597 -0.1897  0.4361  0.1680  0.6513 -0.0433 -1.5741
-1.4514  0.0949 -0.7783  0.8568 -0.8722  0.0364 -0.0998  0.9265
[torch.FloatTensor of size 2x8]

  Tensor维度变型reshaping

x = torch.randn(2, 3, 4)
print(x)
(0 ,.,.) =
  0.6294 -0.3965  1.3737  1.6951
 -0.5477 -1.5385 -0.0288  0.8104
 -0.4208 -0.4469  0.0184  0.9507

(1 ,.,.) =
 -0.2843 -0.0695 -0.1747  2.3774
  1.1067  0.1980 -2.0712 -0.0670
 -1.4900  0.0716 -0.7605  0.4611
[torch.FloatTensor of size 2x3x4]```
view转换维数
  print(x.view(2,12))#将234 -> 2*12 
Columns 0 to 9
0.6294 -0.3965 1.3737 1.6951 -0.5477 -1.5385 -0.0288 0.8104 -0.4208 -0.4469
-0.2843 -0.0695 -0.1747 2.3774 1.1067 0.1980 -2.0712 -0.0670 -1.4900 0.0716

Columns 10 to 11
0.0184 0.9507
-0.7605 0.4611
[torch.FloatTensor of size 2x12]```
print(x.view(2,-1))#-1的话,意味着最后的相乘为维数,这里为2*之后的成绩
#和上面的一样
Columns 0 to 9
 0.6294 -0.3965  1.3737  1.6951 -0.5477 -1.5385 -0.0288  0.8104 -0.4208 -0.4469
-0.2843 -0.0695 -0.1747  2.3774  1.1067  0.1980 -2.0712 -0.0670 -1.4900  0.0716

Columns 10 to 11
 0.0184  0.9507
-0.7605  0.4611
[torch.FloatTensor of size 2x12]

  Computation Graphs and Automatic Differentiation

x = autograd.Variable(torch.Tensor([1., 2., 3]), requires_grad=True)
print(x)
print(x.data)#.data显示具体数据
#找不同
Variable containing:
 1
 2
 3
[torch.FloatTensor of size 3]

 1
 2
 3
[torch.FloatTensor of size 3]
y = autograd.Variable( torch.Tensor([4., 5., 6]), requires_grad=True )
z = x + y
print(z.data)
 5
 7
 9
[torch.FloatTensor of size 3]

  .creator是生成器 

print(z.creator)
#因为是z = x + y 所以,z运算是add
<torch.autograd._functions.basic_ops.Add object at 0x10b04f128>

  显示z中所有元素的和

s = z.sum()
print(s)
Variable containing:
 21
[torch.FloatTensor of size 1]
s.backward()#反向传播
print(x.grad)#对x求导
Variable containing:
 1
 1
 1
[torch.FloatTensor of size 3]
#答案解释
#x = [1,2,3]
#y = [4,5,6]
#z = x + y = [x0+y0, x1+y1, x2+y2]
#s = z.sum() = x0+y0+x1+y1+x2+y2
#x.grad 在s运算中对x求导   也就是当中的x0,x1,x2求导  为1,1,1 

  Deep Learning Building Blocks: Affine maps, non-linearities and objectives

  Affine maps
  也可以说是线性映射,即为f(x) = Ax + b
  nn.Linear(inputSize,outputSize,bias=True)
  输入(N, inputSize)
  输出(N, outputSize)

lin = nn.Linear(5,3)
data = autograd.Variable(torch.randn(2, 5))
print(lin(data))
Variable containing:
-0.1838 -0.1833 -0.6425
 0.2675  0.0263  0.0482
[torch.FloatTensor of size 2x3]

  Non-Linearities
  非线性,常用的函数有 tanh(x),σ(x),ReLU(x) 这些都是激励函数
  在pytorch中大部分激励函数在torch.functional中

data = autograd.Variable( torch.randn(2, 2) )
print(data)
print (F.relu(data))#relu函数是小于零是0,大于零就是它本身
Variable containing:
-2.0620  1.4252
 0.5694  0.2251
[torch.FloatTensor of size 2x2]

Variable containing:
 0.0000  1.4252
 0.5694  0.2251
[torch.FloatTensor of size 2x2]

  Softmax and Probabilities
  softmax是x_i/sum(x)

data = autograd.Variable( torch.randn(5) )
print(data)
print(F.softmax(data))
print(F.softmax(data).sum())
print(F.log_softmax(data))
Variable containing:
 0.6861
 0.1695
-0.4775
-2.0097
 0.7039
[torch.FloatTensor of size 5]

Variable containing:
 0.3340
 0.1992
 0.1043
 0.0225
 0.3400
[torch.FloatTensor of size 5]

Variable containing:
 1
[torch.FloatTensor of size 1]

Variable containing:
-1.0967
-1.6133
-2.2604
-3.7925
-1.0789
[torch.FloatTensor of size 5]

更多精彩内容,请关注 深度学习自然语言处理 公众号,就是下方啦!跟随小博主,每天进步一丢丢!哈哈!

pyTorch基础入门练习-LMLPHP

03-08 06:57