pytorch对CIFAR10图片分类

pytorch官网的教程 https://pytorch.org/tutorials/beginner/blitz/cifar10_tutorial.html

导入cifar10数据集

import torch
import torchvision
import torchvision.transforms as transforms

transform = transforms.Compose(
    [transforms.ToTensor(), #转为tensor
     transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))#归一化
     ])
#训练集
trainset = torchvision.datasets.CIFAR10(root='./data', train=True,
                                        download=True, transform=transform)
# 训练数据加载器
trainloader = torch.utils.data.DataLoader(trainset, batch_size=4,
                                          shuffle=True, num_workers=2)
#测试集
testset = torchvision.datasets.CIFAR10(root='./data', train=False,
                                     download=True, transform=transform)
# 测试数据加载器  
testloader = torch.utils.data.DataLoader(testset, batch_size=4,
                                         shuffle=False, num_workers=2)
# cifar10 分类索引
classes = ('plane', 'car', 'bird', 'cat',
           'deer', 'dog', 'frog', 'horse', 'ship', 'truck')

CIFAR-10中的图像尺寸为3×32×32，即尺寸为32×32像素的3通道彩色图像

dataiter = iter(trainloader)
images, labels = dataiter.next()
images.shape 
torch.Size([64, 3, 32, 32]) B,C,H,W 

显示图片

import matplotlib.pyplot as plt
import numpy as np

# functions to show an image

def imshow(img):
 #输入数据：类型(torch.tensor[c,h,w]：[通道，高度，宽度])
    img = img / 2 + 0.5    #反归一处理
    npimg = img.numpy()
    plt.imshow(np.transpose(npimg, (1, 2, 0))) #图像进行转置，变为[h,w,c]
    plt.show()


# get some random training images
dataiter = iter(trainloader)#加载一个mini batch
images, labels = dataiter.next()

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

定义网络

import torch.nn as nn
import torch.nn.functional as F


class Net(nn.Module):
    def __init__(self):
        super(Net, self).__init__()
        self.conv1 = nn.Conv2d(3, 6, 5)  #输入深度：3   输出深度(卷积核个数)：6   3x3卷积核
        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()

定义损失函数和优化器

import torch.optim as optim

criterion = nn.CrossEntropyLoss()
optimizer = optim.SGD(net.parameters(), lr=0.001, momentum=0.9)

训练网络

for epoch in range(2):  # loop over the dataset multiple times

    running_loss = 0.0
    for i, data in enumerate(trainloader, 0):
        #  输入数据
        inputs, labels = data
		
        # 梯度清零
        optimizer.zero_grad()

        # forward + backward + optimize
        outputs = net(inputs)
        loss = criterion(outputs, labels)
        loss.backward()
        #更新参数
        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')

[1, 2000] loss: 2.208
[1, 4000] loss: 1.887
[1, 6000] loss: 1.702
[1, 8000] loss: 1.607
[1, 10000] loss: 1.530
[1, 12000] loss: 1.473
[2, 2000] loss: 1.410
[2, 4000] loss: 1.371
[2, 6000] loss: 1.324
[2, 8000] loss: 1.311
[2, 10000] loss: 1.316
[2, 12000] loss: 1.289
Finished Training

保存模型

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

取测试集图像测试

# 查看数据,取一组batch
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)))

模型测试

#加载模型
net = Net()
net.load_state_dict(torch.load(PATH))
outputs = net(images)
_, predicted = torch.max(outputs, 1)

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

correct = 0 # 定义预测正确的图片数
total = 0 # 总共参与测试的图片数
with torch.no_grad():
    for data in testloader: # 循环每一个batch
        images, labels = data
        outputs = net(images)
        _, predicted = torch.max(outputs.data, 1)
        total += labels.size(0)   # 更新测试图片的数量
        correct += (predicted == labels).sum().item() #更新正确分类的图片的数量

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

每一类的正确率

class_correct = list(0. for i in range(10))# 每类中测试正确的个数的列表
class_total = list(0. for i in range(10))# 每类中测试总数
with torch.no_grad():# 以batch为单位
    for data in testloader:
        images, labels = data
        outputs = net(images)
        _, predicted = torch.max(outputs, 1)
        c = (predicted == labels).squeeze()
        for i in range(4):# 每个batch都有4张图片
            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]))