PyTorch学习笔记(四) ---- 图像分类器

最新推荐文章于 2024-07-27 10:32:50 发布

john_bh

最新推荐文章于 2024-07-27 10:32:50 发布

阅读量2.5k

点赞数 7

分类专栏： PyTorch 文章标签： PyTorch学习笔记 PyTorch图像分类 Pytorch 图像分类卷积神经网络

本文链接：https://blog.csdn.net/john_bh/article/details/106298979

版权

PyTorch 专栏收录该内容

21 篇文章 41 订阅

订阅专栏

转载请注明作者和出处： http://blog.csdn.net/john_bh/

1.处理数据

对于视觉，已经创建了一个叫做 totchvision 的包，该包含有支持加载类似Imagenet，CIFAR10，MNIST 等公共数据集的数据加载模块 torchvision.datasets 和支持加载图像数据数据转换模块 torch.utils.data.DataLoader。
这次学习将使用CIFAR10数据集，它包含十个类别：‘airplane’, ‘automobile’, ‘bird’, ‘cat’, ‘deer’, ‘dog’, ‘frog’, ‘horse’, ‘ship’, ‘truck’。CIFAR-10 中的图像尺寸为 $3 * 32 * 32$ ，也就是RGB的3层颜色通道，每层通道内的尺寸为 $32 * 32$ 。
在这里插入图片描述
训练一个图像分类器,将按次序的做如下几步：

使用torchvision加载并且归一化CIFAR10的训练和测试数据集
定义一个卷积神经网络
定义一个损失函数
在训练样本数据上训练网络
在测试样本数据上测试网络

2.加载并归一化 CIFAR10

加载并归一化 CIFAR10 使用 torchvision ,用它来加载 CIFAR10 数据非常简单，代码如下：

# -*- coding:utf-8 -*-
import torch
import torchvision
import torchvision.transforms as transforms

# torchvision 数据集的输出是范围在[0,1]之间的 PILImage，我们将他们转换成归一化范围为[-1,1]之间的张量 Tensors。
transform=transforms.Compose([transforms.ToTensor(),transforms.Normalize((0.5,0.5,0.5),(0.5,0.5,0.5))])
trainset=torchvision.datasets.CIFAR10(root="./data",train=True,download=True,transform=transform)
trainloader=torch.utils.data.DataLoader(trainset,batch_size=4,shuffle=True,num_workers=2)

testset=torchvision.datasets.CIFAR10(root="./data",train=False,download=True,transform=transform)
testloader=torch.utils.data.DataLoader(testset,batch_size=4,shuffle=False,num_workers=2)

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

输出：

Files already downloaded and verified
Files already downloaded and verified

因为我已经下载了数据集，所以显示已经下载。

展示其中的一些训练图片。

# -*- coding:utf-8 -*-
import torch
import torchvision
import torchvision.transforms as transforms
import matplotlib.pyplot as plt
import numpy as np

# torchvision 数据集的输出是范围在[0,1]之间的 PILImage，我们将他们转换成归一化范围为[-1,1]之间的张量 Tensors。
transform=transforms.Compose([transforms.ToTensor(),transforms.Normalize((0.5,0.5,0.5),(0.5,0.5,0.5))])
trainset=torchvision.datasets.CIFAR10(root="./data",train=True,download=True,transform=transform)
trainloader=torch.utils.data.DataLoader(trainset,batch_size=4,shuffle=True,num_workers=2)

testset=torchvision.datasets.CIFAR10(root="./data",train=False,download=True,transform=transform)
testloader=torch.utils.data.DataLoader(testset,batch_size=4,shuffle=False,num_workers=2)

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

def imshow(img):
    img=img/2+0.5
    npimg=img.numpy()
    plt.imshow(np.transpose(npimg,(1,2,0)))
    plt.show()
    
dataiter=iter(trainloader)
images,labels=dataiter.next()

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

输出：

Files already downloaded and verified
Files already downloaded and verified

frog horse bird bird

3.定义一个卷积神经网络

随机生成一个 32x32 的输入。注意：期望的输入维度是 32x32 。为了使用这个网络在 MNIST 数据及上，你需要把数据集中的图片维度修改为 32x32。

# -*- coding:utf-8 -*-
import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim

# torchvision 数据集的输出是范围在[0,1]之间的 PILImage，我们将他们转换成归一化范围为[-1,1]之间的张量 Tensors。
transform=transforms.Compose([transforms.ToTensor(),transforms.Normalize((0.5,0.5,0.5),(0.5,0.5,0.5))])
trainset=torchvision.datasets.CIFAR10(root="./data",train=True,download=True,transform=transform)
trainloader=torch.utils.data.DataLoader(trainset,batch_size=4,shuffle=True,num_workers=2)

testset=torchvision.datasets.CIFAR10(root="./data",train=False,download=True,transform=transform)
testloader=torch.utils.data.DataLoader(testset,batch_size=4,shuffle=False,num_workers=2)

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


class Net(nn.Module):
    def __init__(self):
        super(Net,self).__init__()
        self.conv1=nn.Conv2d(3,6,5)
        self.pool=nn.MaxPool2d(2,2)
        self.conv2=nn.Conv2d(6,16,5)
        self.fc1=nn.Linear(16*5*5,120)
        self.fc2=nn.Linear(120,84)
        self.fc3=nn.Linear(84,10)
        
    def forward(self,x):
        x=self.pool(F.relu(self.conv1(x)))
        x=self.pool(F.relu(self.conv2(x)))
        x=x.view(-1,16*5*5)
        x=F.relu(self.fc1(x))
        x=F.relu(self.fc2(x))
        x=self.fc3(x)
        
        return x

net=Net()

4.定义一个损失函数

定义一个损失函数和优化器让我们使用分类交叉熵Cross-Entropy 作损失函数，动量SGD做优化器。

# -*- coding:utf-8 -*-
import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim

# torchvision 数据集的输出是范围在[0,1]之间的 PILImage，我们将他们转换成归一化范围为[-1,1]之间的张量 Tensors。
transform=transforms.Compose([transforms.ToTensor(),transforms.Normalize((0.5,0.5,0.5),(0.5,0.5,0.5))])
trainset=torchvision.datasets.CIFAR10(root="./data",train=True,download=True,transform=transform)
trainloader=torch.utils.data.DataLoader(trainset,batch_size=4,shuffle=True,num_workers=2)

testset=torchvision.datasets.CIFAR10(root="./data",train=False,download=True,transform=transform)
testloader=torch.utils.data.DataLoader(testset,batch_size=4,shuffle=False,num_workers=2)

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


class Net(nn.Module):
    def __init__(self):
        super(Net,self).__init__()
        self.conv1=nn.Conv2d(3,6,5)
        self.pool=nn.MaxPool2d(2,2)
        self.conv2=nn.Conv2d(6,16,5)
        self.fc1=nn.Linear(16*5*5,120)
        self.fc2=nn.Linear(120,84)
        self.fc3=nn.Linear(84,10)
        
    def forward(self,x):
        x=self.pool(F.relu(self.conv1(x)))
        x=self.pool(F.relu(self.conv2(x)))
        x=x.view(-1,16*5*5)
        x=F.relu(self.fc1(x))
        x=F.relu(self.fc2(x))
        x=self.fc3(x)
        
        return x
net=Net()

# 定义一个损失函数和优化器 让我们使用分类交叉熵Cross-Entropy 作损失函数，动量SGD做优化器。
criterion = nn.CrossEntropyLoss()
optimizer=optim.SGD(net.parameters(),lr=0.001,momentum=0.9)

5.在训练样本数据上训练网络

在数据迭代器上循环传给网络和优化器输入进行训练网络，具体代码如下：

# -*- coding:utf-8 -*-
import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim

# torchvision 数据集的输出是范围在[0,1]之间的 PILImage，我们将他们转换成归一化范围为[-1,1]之间的张量 Tensors。
transform=transforms.Compose([transforms.ToTensor(),transforms.Normalize((0.5,0.5,0.5),(0.5,0.5,0.5))])
trainset=torchvision.datasets.CIFAR10(root="./data",train=True,download=True,transform=transform)
trainloader=torch.utils.data.DataLoader(trainset,batch_size=4,shuffle=True,num_workers=2)

testset=torchvision.datasets.CIFAR10(root="./data",train=False,download=True,transform=transform)
testloader=torch.utils.data.DataLoader(testset,batch_size=4,shuffle=False,num_workers=2)

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

class Net(nn.Module):
    def __init__(self):
        super(Net,self).__init__()
        self.conv1=nn.Conv2d(3,6,5)
        self.pool=nn.MaxPool2d(2,2)
        self.conv2=nn.Conv2d(6,16,5)
        self.fc1=nn.Linear(16*5*5,120)
        self.fc2=nn.Linear(120,84)
        self.fc3=nn.Linear(84,10)
        
    def forward(self,x):
        x=self.pool(F.relu(self.conv1(x)))
        x=self.pool(F.relu(self.conv2(x)))
        x=x.view(-1,16*5*5)
        x=F.relu(self.fc1(x))
        x=F.relu(self.fc2(x))
        x=self.fc3(x)
        
        return x
net=Net()

# 定义一个损失函数和优化器 让我们使用分类交叉熵Cross-Entropy 作损失函数，动量SGD做优化器。
criterion = nn.CrossEntropyLoss()
optimizer=optim.SGD(net.parameters(),lr=0.001,momentum=0.9)

# 训练网络:需要在数据迭代器上循环传给网络和优化器 输入就可以。

for epoch in range(2):
    running_loss=0.0
    for i,data in enumerate(trainloader,0):
        inputs,labels=data # get the inputs
        
        optimizer.zero_grad() # zero the parameter gradients
        
        outputs=net(inputs)
        
        # forward + backward + optimize
        loss=criterion(outputs,labels)
        loss.backward()
        optimizer.step()
        
        #print statistics
        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!")

输出：

Files already downloaded and verified
Files already downloaded and verified
[1, 2000] loss: 2.192
[1, 4000] loss: 1.840
[1, 6000] loss: 1.679
[1, 8000] loss: 1.566
[1,10000] loss: 1.503
[1,12000] loss: 1.451
[2, 2000] loss: 1.386
[2, 4000] loss: 1.358
[2, 6000] loss: 1.324
[2, 8000] loss: 1.305
[2,10000] loss: 1.274
[2,12000] loss: 1.240
Finished Training!

6.在测试样本数据上测试网络

已经通过训练数据集对网络进行了2次训练，但是需要检查网络是否已经学到了东西。用神经网络的输出作为预测的类标来检查网络的预测性能，用样本的真实类标来校对。如果预测是正确的，我们将样本添加到正确预测的列表里。

# -*- coding:utf-8 -*-
import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim

# torchvision 数据集的输出是范围在[0,1]之间的 PILImage，我们将他们转换成归一化范围为[-1,1]之间的张量 Tensors。
transform=transforms.Compose([transforms.ToTensor(),transforms.Normalize((0.5,0.5,0.5),(0.5,0.5,0.5))])
trainset=torchvision.datasets.CIFAR10(root="./data",train=True,download=True,transform=transform)
trainloader=torch.utils.data.DataLoader(trainset,batch_size=4,shuffle=True,num_workers=2)

testset=torchvision.datasets.CIFAR10(root="./data",train=False,download=True,transform=transform)
testloader=torch.utils.data.DataLoader(testset,batch_size=4,shuffle=False,num_workers=2)

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

class Net(nn.Module):
    def __init__(self):
        super(Net,self).__init__()
        self.conv1=nn.Conv2d(3,6,5)
        self.pool=nn.MaxPool2d(2,2)
        self.conv2=nn.Conv2d(6,16,5)
        self.fc1=nn.Linear(16*5*5,120)
        self.fc2=nn.Linear(120,84)
        self.fc3=nn.Linear(84,10)
        
    def forward(self,x):
        x=self.pool(F.relu(self.conv1(x)))
        x=self.pool(F.relu(self.conv2(x)))
        x=x.view(-1,16*5*5)
        x=F.relu(self.fc1(x))
        x=F.relu(self.fc2(x))
        x=self.fc3(x)
        
        return x
net=Net()

# 定义一个损失函数和优化器 让我们使用分类交叉熵Cross-Entropy 作损失函数，动量SGD做优化器。
criterion = nn.CrossEntropyLoss()
optimizer=optim.SGD(net.parameters(),lr=0.001,momentum=0.9)

# 训练网络:需要在数据迭代器上循环传给网络和优化器 输入就可以。

for epoch in range(2):
    running_loss=0.0
    for i,data in enumerate(trainloader,0):
        inputs,labels=data # get the inputs
        
        optimizer.zero_grad() # zero the parameter gradients
        
        outputs=net(inputs)
        
        # forward + backward + optimize
        loss=criterion(outputs,labels)
        loss.backward()
        optimizer.step()
        
        #print statistics
        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!") 

outputs = net(images)

# 输出是预测与十个类的近似程度，与某一个类的近似程度越高，网络就越认为图像是属于这一类别。所以让我们打印其中最相似类别类标：
_, predicted = torch.max(outputs, 1)

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

输出：

Files already downloaded and verified
Files already downloaded and verified
[1, 2000] loss: 2.237
[1, 4000] loss: 1.861
[1, 6000] loss: 1.687
[1, 8000] loss: 1.581
[1,10000] loss: 1.533
[1,12000] loss: 1.489
[2, 2000] loss: 1.444
[2, 4000] loss: 1.396
[2, 6000] loss: 1.375
[2, 8000] loss: 1.358
[2,10000] loss: 1.336
[2,12000] loss: 1.324
Finished Training!
Predicted: frog dog frog deer

看看网络在整个数据集上的表现：

# -*- coding:utf-8 -*-
import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim

# torchvision 数据集的输出是范围在[0,1]之间的 PILImage，我们将他们转换成归一化范围为[-1,1]之间的张量 Tensors。
transform=transforms.Compose([transforms.ToTensor(),transforms.Normalize((0.5,0.5,0.5),(0.5,0.5,0.5))])
trainset=torchvision.datasets.CIFAR10(root="./data",train=True,download=True,transform=transform)
trainloader=torch.utils.data.DataLoader(trainset,batch_size=4,shuffle=True,num_workers=2)

testset=torchvision.datasets.CIFAR10(root="./data",train=False,download=True,transform=transform)
testloader=torch.utils.data.DataLoader(testset,batch_size=4,shuffle=False,num_workers=2)

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

class Net(nn.Module):
    def __init__(self):
        super(Net,self).__init__()
        self.conv1=nn.Conv2d(3,6,5)
        self.pool=nn.MaxPool2d(2,2)
        self.conv2=nn.Conv2d(6,16,5)
        self.fc1=nn.Linear(16*5*5,120)
        self.fc2=nn.Linear(120,84)
        self.fc3=nn.Linear(84,10)
        
    def forward(self,x):
        x=self.pool(F.relu(self.conv1(x)))
        x=self.pool(F.relu(self.conv2(x)))
        x=x.view(-1,16*5*5)
        x=F.relu(self.fc1(x))
        x=F.relu(self.fc2(x))
        x=self.fc3(x)
        
        return x
net=Net()

# 定义一个损失函数和优化器 让我们使用分类交叉熵Cross-Entropy 作损失函数，动量SGD做优化器。
criterion = nn.CrossEntropyLoss()
optimizer=optim.SGD(net.parameters(),lr=0.001,momentum=0.9)

# 训练网络:需要在数据迭代器上循环传给网络和优化器 输入就可以。

for epoch in range(2):
    running_loss=0.0
    for i,data in enumerate(trainloader,0):
        inputs,labels=data # get the inputs
        
        optimizer.zero_grad() # zero the parameter gradients
        
        outputs=net(inputs)
        
        # forward + backward + optimize
        loss=criterion(outputs,labels)
        loss.backward()
        optimizer.step()
        
        #print statistics
        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!") 

#网络在整个数据集上的表现
correct = 0
total = 0
with torch.no_grad():
    for data in testloader:
        images, labels = data
        outputs = net(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))

输出：

Files already downloaded and verified
Files already downloaded and verified
[1, 2000] loss: 2.237
[1, 4000] loss: 1.861
[1, 6000] loss: 1.687
[1, 8000] loss: 1.581
[1,10000] loss: 1.533
[1,12000] loss: 1.489
[2, 2000] loss: 1.444
[2, 4000] loss: 1.396
[2, 6000] loss: 1.375
[2, 8000] loss: 1.358
[2,10000] loss: 1.336
[2,12000] loss: 1.324
Finished Training!
Accuracy of the network on the 10000 test images: 53 %

随机预测出为10类中的哪一类，看看模型表现。

# -*- coding:utf-8 -*-
import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim

# torchvision 数据集的输出是范围在[0,1]之间的 PILImage，我们将他们转换成归一化范围为[-1,1]之间的张量 Tensors。
transform=transforms.Compose([transforms.ToTensor(),transforms.Normalize((0.5,0.5,0.5),(0.5,0.5,0.5))])
trainset=torchvision.datasets.CIFAR10(root="./data",train=True,download=True,transform=transform)
trainloader=torch.utils.data.DataLoader(trainset,batch_size=4,shuffle=True,num_workers=2)

testset=torchvision.datasets.CIFAR10(root="./data",train=False,download=True,transform=transform)
testloader=torch.utils.data.DataLoader(testset,batch_size=4,shuffle=False,num_workers=2)

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

class Net(nn.Module):
    def __init__(self):
        super(Net,self).__init__()
        self.conv1=nn.Conv2d(3,6,5)
        self.pool=nn.MaxPool2d(2,2)
        self.conv2=nn.Conv2d(6,16,5)
        self.fc1=nn.Linear(16*5*5,120)
        self.fc2=nn.Linear(120,84)
        self.fc3=nn.Linear(84,10)
        
    def forward(self,x):
        x=self.pool(F.relu(self.conv1(x)))
        x=self.pool(F.relu(self.conv2(x)))
        x=x.view(-1,16*5*5)
        x=F.relu(self.fc1(x))
        x=F.relu(self.fc2(x))
        x=self.fc3(x)
        
        return x
net=Net()

# 定义一个损失函数和优化器 让我们使用分类交叉熵Cross-Entropy 作损失函数，动量SGD做优化器。
criterion = nn.CrossEntropyLoss()
optimizer=optim.SGD(net.parameters(),lr=0.001,momentum=0.9)

# 训练网络:需要在数据迭代器上循环传给网络和优化器 输入就可以。

for epoch in range(2):
    running_loss=0.0
    for i,data in enumerate(trainloader,0):
        inputs,labels=data # get the inputs
        
        optimizer.zero_grad() # zero the parameter gradients
        
        outputs=net(inputs)
        
        # forward + backward + optimize
        loss=criterion(outputs,labels)
        loss.backward()
        optimizer.step()
        
        #print statistics
        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!") 

#随机预测出为10类中的哪一类，看来网络性能
class_correct = list(0. for i in range(10))
class_total = list(0. for i in range(10))
with torch.no_grad():
    for data in testloader:
        images, labels = data
        outputs = net(images)
        _, predicted = torch.max(outputs, 1)
        c = (predicted == labels).squeeze()
        for i in range(4):
            label = labels[i]
            class_correct[label] += c[i].item()
            class_total[label] += 1


for i in range(10):
    print('Accuracy of %5s : %2d %%' % (
        classes[i], 100 * class_correct[i] / class_total[i]))

输出：

Files already downloaded and verified
Files already downloaded and verified
[1, 2000] loss: 2.237
[1, 4000] loss: 1.861
[1, 6000] loss: 1.687
[1, 8000] loss: 1.581
[1,10000] loss: 1.533
[1,12000] loss: 1.489
[2, 2000] loss: 1.444
[2, 4000] loss: 1.396
[2, 6000] loss: 1.375
[2, 8000] loss: 1.358
[2,10000] loss: 1.336
[2,12000] loss: 1.324
Finished Training!
Accuracy of plane : 57 %
Accuracy of car : 51 %
Accuracy of bird : 23 %
Accuracy of cat : 17 %
Accuracy of deer : 46 %
Accuracy of dog : 58 %
Accuracy of frog : 80 %
Accuracy of horse : 58 %
Accuracy of ship : 76 %
Accuracy of truck : 67 %