PyTorch官方教程 - Getting Started - 60分钟快速入门 - 训练分类器

TRAINING A CLASSIFIER

数据加载

torchvision

torchvision.datasets

torch.utils.data.DataLoader

训练图像分类器

1. 使用torchvision加载及标准化CIFAR10训练、测试数据集
2. 定义卷积神经网络
3. 定义损失函数
4. 训练网络
5. 测试网络

1. 加载及标准化CIFAR10训练、测试数据集

import torch
import torchvision
import torchvision.transforms as transforms

torchvision数据集输出为PILImage图像，其取值范围为$[0,1]$，将其转化为取值范围为$[-1,1]$的标准化张量。

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

trainset = torchvision.datasets.CIFAR10(root="../../../datasets", 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="../../../datasets", 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")

Files already downloaded and verified
Files already downloaded and verified

%matplotlib inline
import matplotlib.pyplot as plt
import numpy as np

# functions to show an image
def imshow(img):
    img = img / 2 + 0.5     # unnormalize
    npimg = img.numpy()
    plt.imshow(np.transpose(npimg, (1, 2, 0)))
    plt.show()
    
# get some random training images
dataiter = iter(trainloader)
images, labels = dataiter.next()

# show images
imshow(torchvision.utils.make_grid(images))
# print labels
print(" ".join(["{:5s}".format(classes[labels[j]]) for j in range(4)]))

plane ship  deer  ship 

2 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(in_channels=3, out_channels=6, kernel_size=5)
        self.pool = nn.MaxPool2d(kernel_size=2, stride=2)
        self.conv2 = nn.Conv2d(in_channels=6, out_channels=16, kernel_size=5)
        self.fc1 = nn.Linear(in_features=16 * 5 * 5, out_features=120)
        self.fc2 = nn.Linear(in_features=120, out_features=84)
        self.fc3 = nn.Linear(in_features=84, out_features=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=(5, 5), 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)
)

3 定义损失函数

import torch.optim as optim

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

4 训练网络

for epoch in range(2):  # loop over the dataset multiple times
    
    running_loss = 0
    for i, data in enumerate(trainloader, start=0):
        # get the inputs; data is a list of [inputs, labels]
        inputs, labels = data
        
        # zero the parameter gradients
        optimizer.zero_grad()
        
        # forward + backward + optimize
        outputs = net(inputs)
        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}"
                  .format(epoch + 1, i + 1, running_loss / 2000))
            running_loss = 0
            
            
print("Finishe Training")

[1,  2000] loss: 2.225
[1,  4000] loss: 1.892
[1,  6000] loss: 1.692
[1,  8000] loss: 1.583
[1, 10000] loss: 1.506
[1, 12000] loss: 1.482
[2,  2000] loss: 1.402
[2,  4000] loss: 1.367
[2,  6000] loss: 1.338
[2,  8000] loss: 1.330
[2, 10000] loss: 1.311
[2, 12000] loss: 1.308
Finishe Training

5 测试网络

dataiter = iter(testloader)
images, labels = dataiter.next()

# print images
imshow(torchvision.utils.make_grid(images))
print("GroundTruth: {}".format(
    " ".join(["{:5s}".format(classes[labels[j]]) for j in range(4)])))

GroundTruth: cat   ship  ship  plane

outputs = net(images)
_, predicted = torch.max(outputs, dim=1)
print("Predicted: {}".format(" ".join(["{:5s}".format(classes[predicted[j]])
                                       for j in range(4)])))

Predicted: cat   ship  ship  ship 

correct = 0
total = 0
with torch.no_grad():
    for data in testloader:
        images, labels = data
        outputs = net(images)
        _, predicted = torch.max(outputs, dim=1)
        total += labels.size(0)
        correct += (predicted == labels).sum().item()
        
print("Accuracy of the network on the 10000 test images: {}%".format(100 * correct / total))

Accuracy of the network on the 10000 test images: 55.81%

class_correct = [0] * 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, dim=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} %".format(
        classes[i], 100 * class_correct[i] // class_total[i]))

Accuracy of plane : 59 %
Accuracy of car   : 75 %
Accuracy of bird  : 40 %
Accuracy of cat   : 17 %
Accuracy of deer  : 32 %
Accuracy of dog   : 63 %
Accuracy of frog  : 70 %
Accuracy of horse : 59 %
Accuracy of ship  : 79 %
Accuracy of truck : 59 %

使用GPU训练

查询GPU是否可用

device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")

# Assuming that we are on a CUDA machine, this should print a CUDA device:
print(device)

cuda:0

将网络参数和缓存转为CUDA张量

net.to(device)

Net(
  (conv1): Conv2d(3, 6, kernel_size=(5, 5), 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)
)

训练中，每一步都需将输入和目标转为CUDA张量

inputs, labels = data[0].to(device), data[1].to(device)

使用多GPU训练

删除数据迭代器

del dataiter