pytorch学习笔记__分类器官方代码解读

具体函数使用方法参考官方文档
https://pytorch-cn.readthedocs.io/zh/latest/torchvision/torchvision-datasets/

1、数据转化

transform = transforms.Compose(
    [transforms.ToTensor(), 
     transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))])  
    #把一个取值范围是[0,255]的PIL.Image 转换成 Tensor,将取值变换到（-1,1）之间

trainset = torchvision.datasets.CIFAR10(root='./data', train=True, download=True, transform=transform) 
#载入训练集
trainloader = torch.utils.data.DataLoader(trainset, batch_size=4, shuffle=True, num_workers=2) 
#每个batch加载4个样本，每个epoch重新打乱数据
testset =   torchvision.datasets.CIFAR10(root='./data', train=False, download=True, transform=transform)  
#载入测试集
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')

2、训练集的图片与标签

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)))
# 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' % classes[labels[j]] for j in range(4)))

3、定义网络

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

class Net(nn.Module):
    def __init__(self):  #定义层的参数
        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):  #定义数据在层之间的流动顺序
        x = self.pool(F.relu(self.conv1(x)))
        x = self.pool(F.relu(self.conv2(x)))
        x = x.view(-1, 16 * 5 * 5)   #变成一维张量，接入全连接层
        x = F.relu(self.fc1(x))
        x = F.relu(self.fc2(x))
        x = self.fc3(x)
        return x
        
net = Net()

网络结构如下

Net(
  (conv1): Conv2d(3, 6, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2))
  (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), padding=(2, 2))
  (conv3): Conv2d(16, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
  (conv4): Conv2d(32, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
  (fc1): Linear(in_features=512, out_features=120, bias=True)
  (fc2): Linear(in_features=120, out_features=10, bias=True)
)

4、损失函数

https://pytorch-cn.readthedocs.io/zh/latest/package_references/torch-optim/

import torch.optim as optim

criterion = nn.CrossEntropyLoss() #交叉熵损失函数
optimizer = optim.SGD(net.parameters(), lr=0.001, momentum=0.9) #随机梯度下降，学习率0.001，动量因子0.9

损失函数用的是随机梯度下降。
例子的学习率是固定的，在实践中，有必要随着时间的推移逐渐降低学习率。

class torch.optim.SGD(params, lr=, momentum=0, dampening=0, weight_decay=0, nesterov=False)[source]

参数：
params (iterable) – 待优化参数的iterable或者是定义了参数组的dict
lr (float) – 学习率
momentum (float, 可选) – 动量因子（默认：0）
weight_decay (float, 可选) – 权重衰减（L2惩罚）（默认：0）
dampening (float, 可选) – 动量的抑制因子（默认：0）
nesterov (bool, 可选) – 使用Nesterov动量（默认：False）

5、训练过程

for epoch in range(2):  # 全部训练集训练两次：epoch=[0,1]

    running_loss = 0.0  #清空
    for i, data in enumerate(trainloader, 0):
        # get the inputs
        inputs, labels = data
        # zero the parameter gradients
        optimizer.zero_grad()
        # forward + backward + optimize
        outputs = net(inputs)
        loss = criterion(outputs, labels) #计算损失
        loss.backward()  #反向传递
        optimizer.step() #更新所有参数
        # print statistics
        running_loss += loss.item()  #计算每次batch的损失
        
        if i % 2000 == 1999:    # print every 2000 mini-batches  #每2000次batch计算一次loss
            print('[%d, %5d] loss: %.3f' %
                  (epoch + 1, i + 1, running_loss / 2000))
            running_loss = 0.0

print('Finished Training')

6、测试过程

dataiter = iter(testloader) #迭代
images, labels = dataiter.next() #配套迭代使用效果更佳

# print images 随意测试四张测试集图片看准确度
imshow(torchvision.utils.make_grid(images))
print('GroundTruth: ', ' '.join('%5s' % classes[labels[j]] for j in range(4)))

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

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

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))

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]))

7、通过GPU训练

device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
# Assume that we are on a CUDA machine, then this should print a CUDA device:
print(device)
#cpu移到GPU主要移俩：数据和网络
net.to(device)  #移网络
inputs, labels = inputs.to(device),labels.to(device)  #移数据