Pytorch_訓練簡單分類器

訓練分類器

訓練一個圖像分類器

1. 使用torchvision加載和歸一化CIFAR10訓練集和測試集
2. 定義一個卷積神經網絡
3. 定義損失函數
4. 在訓練集上訓練網絡
5. 在測試集上測試網絡

torchvision輸出的是[0,1]的PILImage圖像，歸一化範圍是[-1, 1]的張量

%matplotlib inline

定義損失函數和優化器
損失函數使用交叉熵，使用隨機梯度下降

import torch
import torchvision
from torchvision import transforms
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim


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


class Run(object):
    '''docstring for Run'''

    def __init__(self):
        super(Run, self).__init__()

    def train(self, epoch):
        model.train()
        criterion = nn.CrossEntropyLoss()
        optimizer = optim.SGD(model.parameters(), lr=0.01, momentum=0.9)
        for epoch in range(epoch):
            running_loss = 0
            for i, data in enumerate(trainloader, 0):
                inputs, labels = data
                optimizer.zero_grad()
                outputs = model(inputs)
                loss = criterion(outputs, labels)
                loss.backward()
                optimizer.step()

                # 打印狀態信息
                running_loss += loss.item()
                if i % 2000 == 0:
                    print('[%d, %5d] loss: %.3f' % (epoch + 1, i + 1, running_loss / 2000))
                    running_loss = 0
        print('training finshed')


if __name__ == '__main__':
    transform = transforms.Compose([transforms.ToTensor(),
                                    transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))])

    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')
    model = Net()
    r = Run()
    r.train(2)

Files already downloaded and verified
Files already downloaded and verified
[1,     1] loss: 0.001
[1,  2001] loss: 2.090
[1,  4001] loss: 1.978
[1,  6001] loss: 1.981
[1,  8001] loss: 2.001
[1, 10001] loss: 1.973
[1, 12001] loss: 1.988
[2,     1] loss: 0.001
[2,  2001] loss: 2.010
[2,  4001] loss: 1.989
[2,  6001] loss: 2.014
[2,  8001] loss: 2.040
[2, 10001] loss: 2.011
[2, 12001] loss: 2.032
training finshed