训练分类器

前言

最近几天都在学习pyTorch，深度学习入门的确很难，下面是自己学习pytorch写的一些代码注释，请大家多多支持

加载并标准化CIFAR10

transform = transforms.Compose(
    [transforms.ToTensor(),
     transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))])#用均值和标准偏差对张量图像进行标准化
     #前一个括号为mean，后一个括号是std,image=(image-mean)/std
     #前面的（0.5，0.5，0.5） 是 R G B 三个通道上的均值， 后面(0.5, 0.5, 0.5)是三个通道的标准差，
#torchvision.transforms是pytorch中的图像预处理包。一般用Compose把多个步骤整合到一起
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')

展示图片

import matplotlib.pyplot as plt
import numpy as np

#functions to show an image

def imshow(img):
    img = img / 2 + 0.5     # unnormalize上一步的标准化的逆变换，返回原来的图片
    npimg = img.numpy()#将torch.FloatTensor 转换为numpy
    plt.imshow(np.transpose(npimg, (1, 2, 0)))
    '''因为在plt.imshow在现实的时候输入的是（imagesize,imagesize,channels）
    而def imshow(img,text,should_save=False)中
    参数img的格式为（channels,imagesize,imagesize）
    这两者的格式不一致，我们需要调用一次np.transpose函数，
    即np.transpose(npimg,(1,2,0))，
    将npimg的数据格式由（channels,imagesize,imagesize）
    转化为（imagesize,imagesize,channels）,进行格式的转换后方可进行显示。'''
    plt.show()
#get some random training images
dataiter = iter(trainloader)
images, labels = dataiter.next()
#打包后的数据集使用 iter() 获取一个批次的数据。

#show images
imshow(torchvision.utils.make_grid(images))
#make_grid的作用是将若干幅图像拼成一幅图像。

#print labels
print(' '.join('%5s' % classes[labels[j]] for j in range(4)))
#python字符串格式化 ' '.join表示用空格来连接后面的字符串

定义卷积神经网络

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)
        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(使用分类交叉熵损失和带有动量的SGD)

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):
        # get the inputs; data is a list of [inputs, labels]
        inputs, labels = data

        # zero the parameter gradients
        optimizer.zero_grad()#把梯度置零，也就是把loss关于weight的导数变成0

        # forward + backward + optimize
        outputs = net(inputs)#定义网络
        loss = criterion(outputs, labels)#定义loss: criterion = nn.CrossEntropyLoss()
        loss.backward()#自动求导
        optimizer.step()
        #用来更新参数的,定义优化方法： optimizer = torch.optim.SGD(net.parameters(), lr=0.001, momentum=0.9)

        # print statistics
        running_loss += loss.item()
        if i % 2000 == 1999:    # print every 2000 mini-batches,每个2000次之类先用running_loss进行累加
            print('[%d, %5d] loss: %.3f' %
                  (epoch + 1, i + 1, running_loss / 2000))#然后再除以2000，就得到这两千次的平均损失值
            running_loss = 0.0# # 这一个2000次结束后，就把running_loss归零，下一个2000次继续使用

print('Finished Training')

在测试数据上测试网络

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)))

net = Net()
outputs = net(images)
_, predicted = torch.max(outputs, 1)

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

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))

分类看

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])