Pytorch: Training a Classifier (四)

这个还是官网 tutorial上的一个例子，还是非常有意思的，感觉搭建网络不用一开始就上大数据集，可以先用小数据集进行简单训练和inference，为啥呢？

如果上来一个大数据集，每次调试载入数据都得30分钟，然后发现存在一个小问题，还不如用个小数据集，确认无误了，再用大数据集，包括预处理也是，先用一部分数据测试可行性，不然时间成本太高了。。。

#载入对应依赖库
import torch
import torchvision
import torchvision.transforms as transforms

#官网上说数据集的输出为[0,1]的范围
#要将其的转化为范围为[-1, 1]
#因为输入数据是3个通道的，所以设置每个通道的mean为（0.5,0.5,0.5），
# standard deviations为（0.5，0.5,0.5）
transform = transforms.Compose([transforms.ToTensor(),
                               transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))])

#这边建议download=False,自己到网站上去下载对应的数据到./data目录下，不然的话会很久
#有种错觉是不是代码写错了。。。
trainset = torchvision.datasets.CIFAR10(root='./data', train=True, download=False, transform = transform)
trainloader = torch.utils.data.DataLoader(trainset, batch_size=4, shuffle=True, num_workers = 0)
testset = torchvision.datasets.CIFAR10(root='./data', train= False, download=False, transform = transform)
testloader = torch.utils.data.DataLoader(testset, batch_size=4, shuffle=False, num_workers = 0)

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

#显示一些图片
import matplotlib.pyplot as plt
import numpy as np

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()


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

      deer     truck      dog      cat

#开始构建自己的网络
import torch
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, 3)
        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()
print(net)

Net(
  (conv1): Conv2d(3, 6, kernel_size=(3, 3), stride=(1, 1))
  (pool): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
  (conv2): Conv2d(6, 16, kernel_size=(5, 5), stride=(1, 1))
  (fc1): Linear(in_features=400, out_features=120, bias=True)
  (fc2): Linear(in_features=120, out_features=84, bias=True)
  (fc3): Linear(in_features=84, out_features=10, bias=True)
)

#设置交叉熵作为损失函数，不清楚的可以看cs231n的课程
import torch.optim as optim
criterion = nn.CrossEntropyLoss()
optimizer = optim.SGD(net.parameters(), lr=0.001, momentum=0.9)

#设置epoch为2，也就是整个数据学习两轮
for epoch in range(2):
    running_loss = 0.0
    for i, data in enumerate(trainloader, 0):
        inputs, labels = data
        optimizer.zero_grad()
        
        outputs = net(inputs)
        loss = criterion(outputs, labels)
        loss.backward()
        optimizer.step()
        
        running_loss += loss.item()
        
        if i % 2000 == 1999:
            print('[%d, %5d] loss: %.3f' %(epoch+1, i+1, running_loss/2000))
            running_loss = 0.0
print('Finished Training')

[1,  2000] loss: 1.229
[1,  4000] loss: 1.234
[1,  6000] loss: 1.210
[1,  8000] loss: 1.219
[1, 10000] loss: 1.206
[1, 12000] loss: 1.209
[2,  2000] loss: 1.134
[2,  4000] loss: 1.145
[2,  6000] loss: 1.136
[2,  8000] loss: 1.119
[2, 10000] loss: 1.138
[2, 12000] loss: 1.117
Finished Training

#保存模型
PATH = './cifar_net.pth'
torch.save(net.state_dict(), PATH)

#显示部分测试的图片
dataiter = iter(testloader)
images, labels = dataiter.next()

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

GroundTruth:    cat    ship     ship    plane

#重新加载模型
net = Net()
net.load_state_dict(torch.load(PATH))

IncompatibleKeys(missing_keys=[], unexpected_keys=[])

#预测一下，好像对了50%。。。。
_, predicted = torch.max(outputs, 1)
print('Predicted: ', ' '.join('%5s'% classes[predicted[j]] for j in range(4)))

Predicted:   frog   ship    car    plane

#测试总的准确率
correct = 0
total = 0
with torch.no_grad():
    for data in testloader:
        images, labels = data
        outputs = net(images)
        #输出可能性最大的index
        _, predicted = torch.max(outputs.data, 1)
        total += labels.size(0)
        #计算label和predicted相同的个数总和，就是预测对的数量
        correct += (predicted == labels).sum().item()
        
print('Accuracy of the network on the 10000 test images: %d %%' %(100*correct /total))

Accuracy of the network on the 10000 test images: 57 %

#查看具体每一类的准确率
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]))

猫和狗学的好差。。。。
Accuracy of plane : 80 %
Accuracy of   car : 79 %
Accuracy of  bird : 37 %
Accuracy of   cat : 33 %
Accuracy of  deer : 44 %
Accuracy of   dog : 39 %
Accuracy of  frog : 63 %
Accuracy of horse : 58 %
Accuracy of  ship : 76 %
Accuracy of truck : 60 %

#如果有GPU，当然使用GPU啦，不然好浪费，感觉LZ的970m小笔记本都快被榨干了。。。
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
print(device)

cuda:0

#将网络和数据从CPU上移到GPU上去
net.to(device)
inputs, labels = data[0].to(device), data[1].to(device)

感觉基本的操作都进行了一遍，后续还要继续学习O(∩_∩)O哈哈~

参考地址：
https://pytorch.org/tutorials/beginner/blitz/cifar10_tutorial.html#sphx-glr-beginner-blitz-cifar10-tutorial-py