pytorch入门

"1"
PyTorch是一个基于Python的库,提供了一个具有灵活易用的深度学习框架,是近年来最受欢迎的深度学习框架之一。

初识PyTorch

张量

1.导入pytorch包

1
import torch

2.创建一个空的5x3张量

1
2
x = torch.empty(5, 3)
print(x)

3.创建一个随机初始化的5x3张量

1
2
x = torch.rand(5, 3)
print(x)

4.创建一个5x3的0张量,类型为long

1
2
x = torch.zeros(5, 3, dtype=torch.long)
print(x)

5.直接从数组创建张量

1
2
x = torch.tensor([5.5, 3])
print(x)

6.创建一个5x3的单位张量,类型为double

1
2
x = torch.ones(5, 3, dtype=torch.double)
print(x)

7.从已有的张量创建相同维度的新张量,并且重新定义类型为float

1
2
x = torch.randn_like(x, dtype=torch.float)
print(x)

8.打印一个张量的维度

1
print(x.size())

9.将两个张量相加

1
2
3
4
5
6
7
8
9
10
11
12
13
14
y = torch.rand(5, 3)
print(x + y)

# 方法二
# print(torch.add(x, y))

# 方法三
# result = torch.empty(5, 3)
# torch.add(x, y, out=result)
# print(result)

# 方法四
# y.add_(x)
# print(y)

10.取张量的第一列

1
print(x[:, 1])

11.将一个4x4的张量resize成一个一维张量

1
2
3
x = torch.randn(4, 4)
y = x.view(16)
print(x.size(),y.size())

12.将一个4x4的张量,resize成一个2x8的张量

1
2
3
4
5
6
y = x.view(2, 8)
print(x.size(),y.size())

# 方法二
z = x.view(-1, 8) # 确定一个维度,-1的维度会被自动计算
print(x.size(),z.size())

13.从张量中取出数字x = torch.randn(1)

1
2
print(x)
print(x.item())

Numpy的操作

14.将张量装换成numpy数组

1
2
3
4
5
a = torch.ones(5)
print(a)

b = a.numpy()
print(b)

15.将张量+1,并观察上题中numpy数组的变化

1
2
3
a.add_(1)
print(a)
print(b)

16.从numpy数组创建张量

1
2
3
4
5
import numpy as np
a = np.ones(5)
b = torch.from_numpy(a)
print(a)
print(b)

17.将numpy数组+1并观察上题中张量的变化

1
2
3
np.add(a, 1, out=a)
print(a)
print(b)

自动微分

张量的自动微分

18.新建一个张量,并设置requires_grad=True

1
2
x = torch.ones(2, 2, requires_grad=True)
print(x)

19.对张量进行任意操作(y = x + 2)

1
2
3
y = x + 2
print(y)
print(y.grad_fn) # y就多了一个AddBackward

20.再对y进行任意操作

1
2
3
4
5
z = y * y * 3
out = z.mean()

print(z) # z多了MulBackward
print(out) # out多了MeanBackward

梯度

21.对out进行反向传播

1
out.backward()

22.打印梯度d(out)/dx

1
print(x.grad) #out=0.25*Σ3(x+2)^2

23.创建一个结果为矢量的计算过程(y=x*2^n)

1
2
3
4
5
6
7
x = torch.randn(3, requires_grad=True)

y = x * 2
while y.data.norm() < 1000:
y = y * 2

print(y)

24.计算v = [0.1, 1.0, 0.0001]处的梯度

1
2
3
4
v = torch.tensor([0.1, 1.0, 0.0001], dtype=torch.float)
y.backward(v)

print(x.grad)

25.关闭梯度的功能

1
2
3
4
5
6
7
8
9
10
11
print(x.requires_grad)
print((x ** 2).requires_grad)

with torch.no_grad():
print((x ** 2).requires_grad)

# 方法二
# print(x.requires_grad)
# y = x.detach()
# print(y.requires_grad)
# print(x.eq(y).all())

神经网络

这部分以LeNet5为例,
"2"

定义网络

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
import torch
import torch.nn as nn
import torch.nn.functional as F

class Net(nn.Module):

def __init__(self):
super(Net, self).__init__()
# 26.定义①的卷积层,输入为32x32的图像,卷积核大小5x5卷积核种类6
self.conv1 = nn.Conv2d(3, 6, 5)
# 27.定义③的卷积层,输入为前一层6个特征,卷积核大小5x5,卷积核种类16
self.conv2 = nn.Conv2d(6, 16, 5)
# 28.定义⑤的全链接层,输入为16*5*5,输出为120
self.fc1 = nn.Linear(16 * 5 * 5, 120) # 6*6 from image dimension
# 29.定义⑥的全连接层,输入为120,输出为84
self.fc2 = nn.Linear(120, 84)
# 30.定义⑥的全连接层,输入为84,输出为10
self.fc3 = nn.Linear(84, 10)

def forward(self, x):
# 31.完成input-S2,先卷积+relu,再2x2下采样
x = F.max_pool2d(F.relu(self.conv1(x)), (2, 2))
# 32.完成S2-S4,先卷积+relu,再2x2下采样
x = F.max_pool2d(F.relu(self.conv2(x)), 2) #卷积核方形时,可以只写一个维度
# 33.将特征向量扁平成行向量
x = x.view(-1, 16 * 5 * 5)
# 34.使用fc1+relu
x = F.relu(self.fc1(x))
# 35.使用fc2+relu
x = F.relu(self.fc2(x))
# 36.使用fc3
x = self.fc3(x)
return x


net = Net()
print(net)

37.打印网络的参数

1
2
3
params = list(net.parameters())
# print(params)
print(len(params))

38.打印某一层参数的形状

1
print(params[0].size())

39.随机输入一个向量,查看前向传播输出

1
2
3
input = torch.randn(1, 1, 32, 32)
out = net(input)
print(out)

40.将梯度初始化net.zero_grad()
41.随机一个梯度进行反向传播

1
out.backward(torch.randn(1, 10))

损失函数

42.用自带的MSELoss()定义损失函数

1
criterion = nn.MSELoss()

43.随机一个真值,并用随机的输入计算损失

1
2
3
4
5
6
7
target = torch.randn(10)  # 随机真值
target = target.view(1, -1) # 变成行向量

output = net(input) # 用随机输入计算输出

loss = criterion(output, target) # 计算损失
print(loss)

44.将梯度初始化,计算上一步中loss的反向传播

1
2
3
4
net.zero_grad()

print('conv1.bias.grad before backward')
print(net.conv1.bias.grad)

45.计算43中loss的反向传播

1
2
3
4
loss.backward()

print('conv1.bias.grad after backward')
print(net.conv1.bias.grad)

更新权重

46.定义SGD优化器算法,学习率设置为0.01

1
2
import torch.optim as optim
optimizer = optim.SGD(net.parameters(), lr=0.01)

47.使用优化器更新权重

1
2
3
4
5
6
7
optimizer.zero_grad()
output = net(input)
loss = criterion(output, target)
loss.backward()

# 更新权重
optimizer.step()

训练一个分类器

读取CIFAR10数据,做标准化

48.构造一个transform,将三通道(0,1)区间的数据转换成(-1,1)的数据

1
2
3
4
5
6
import torchvision
import torchvision.transforms as transforms

transform = transforms.Compose(
[transforms.ToTensor(),
transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))])

读取数据集:

1
2
3
4
5
6
7
trainset = cifar(root = './input/cifar10', segmentation='train', transforms=transform)
testset = cifar(root = './input/cifar10', segmentation='test', transforms=transform)
trainloader = torch.utils.data.DataLoader(trainset, batch_size=batch_size,shuffle=True, num_workers=2)
testloader = torch.utils.data.DataLoader(testset, batch_size=batch_size,shuffle=False, num_workers=2)

classes = ('plane', 'car', 'bird', 'cat',
'deer', 'dog', 'frog', 'horse', 'ship', 'truck')

建立网络

使用前面的网络:

1
net2 = Net()

定义损失函数和优化器

49.定义交叉熵损失函数

1
criterion2 = nn.CrossEntropyLoss()

50.定义SGD优化器算法,学习率设置为0.001,momentum=0.9

1
optimizer2 = optim.SGD(net2.parameters(), lr=0.001, momentum=0.9)

训练网络

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
作者:一两赘肉无
链接:https://zhuanlan.zhihu.com/p/99318332
来源:知乎
著作权归作者所有。商业转载请联系作者获得授权,非商业转载请注明出处。

for epoch in range(2):

running_loss = 0.0
for i, data in enumerate(trainloader, 0):
# 获取X,y对
inputs, labels = data

# 51.初始化梯度
optimizer2.zero_grad()

# 52.前馈
outputs = net2(inputs)
# 53.计算损失
loss = criterion2(outputs, labels)
# 54.计算梯度
loss.backward()
# 55.更新权值
optimizer2.step()

# 每2000个数据打印平均代价函数值
running_loss += loss.item()
if i % 2000 == 1999: # print every 2000 mini-batches
print('[%d, %5d] loss: %.3f' %
(epoch + 1, i + 1, running_loss / 2000))
running_loss = 0.0

print('Finished Training')

使用模型预测

读一些数据

1
2
3
4
5
6
dataiter = iter(testloader)
images, labels = dataiter.next()

# print images
imshow(torchvision.utils.make_grid(images))
print('GroundTruth: ', ' '.join('%5s' % classes[labels[j]] for j in range(4)))

56.使用模型预测

1
2
3
4
5
6
outputs = net2(images)

_, predicted = torch.max(outputs, 1)

print('Predicted: ', ' '.join('%5s' % classes[predicted[j]]
for j in range(4)))

57.在测试集上进行打分

1
2
3
4
5
6
7
8
9
10
11
12
correct = 0
total = 0
with torch.no_grad():
for data in testloader:
images, labels = data
outputs = net2(images)
_, predicted = torch.max(outputs.data, 1)
total += labels.size(0)
correct += (predicted == labels).sum().item()

print('Accuracy of the network on the 10000 test images: %d %%' % (
100 * correct / total))

存取模型

58.保存训练好的模型

1
2
PATH = './cifar_net.pth'
torch.save(net.state_dict(), PATH)

59.读取保存的模型

1
pretrained_net = torch.load(PATH)

60.加载模型

1
2
3
net3 = Net()

net3.load_state_dict(pretrained_net)