卷积神经网络识别mnist手写数据集

卷积神经网络识别mnist手写数据集

import torch
import torch.nn.functional as F
from torchvision import transforms
from torchvision import datasets
from torch.utils.data import DataLoader
import torch.optim as optim
import matplotlib.pyplot as plt

batch_size = 4
#转变数形
transform = transforms.Compose([
    transforms.ToTensor(),
    transforms.Normalize((0.1307, ),(0.3081, )) ])
#加载数据
train_dataset = datasets.MNIST(root='../data/j/', train=True, download=True, transform=transform)
train_loader = DataLoader(train_dataset, shuffle=True, batch_size=batch_size)
test_dataset = datasets.MNIST(root='../data/j/', train=False, download=True, transform=transform)
test_loader = DataLoader(test_dataset, shuffle=False, batch_size=batch_size)

'''cnn网络模型'''
class Net(torch.nn.Module):
    def __init__(self):
        super(Net, self).__init__()
        self.conv1 = torch.nn.Conv2d(1, 10, kernel_size=5)
        self.conv2 = torch.nn.Conv2d(10, 20, kernel_size=5)
        self.pooling = torch.nn.MaxPool2d(2)
        self.fc = torch.nn.Linear(320, 10)
    def forward(self, x):
        # Flatten data from (n, 1, 28, 28) to (n, 784)
         batch_size = x.size(0)
         x = F.relu(self.pooling(self.conv1(x)))
         x = F.relu(self.pooling(self.conv2(x)))
         x = x.view(batch_size, -1) # flatten
         x = self.fc(x)
         return x


net1 = Net()
#损失函数
criterion = torch.nn.CrossEntropyLoss()
#优化器
optimizer = optim.SGD(net1.parameters(), lr=0.01, momentum=0.5)

def train(epoch):
    running_loss = 0.0
    for batch_idx, data in enumerate(train_loader, 0):
        inputs, target = data
        optimizer.zero_grad()
        # forward + backward + update
        outputs = net1(inputs)
        loss = criterion(outputs, target)
        loss.backward()
        optimizer.step()
        running_loss += loss.item()
        if batch_idx % 300 == 299:
            print('[%d, %5d] loss: %.3f' % (epoch + 1, batch_idx + 1, running_loss / 2000))
            running_loss = 0.0

def imshow(inputs, picname):
    plt.ion()
    inputs = inputs / 2 + 0.5  # 加载数据集的时候使用了正则化，需要恢复一下
    inputs = inputs.numpy().transpose((1, 2, 0))  # pytorch加载图片通道数在前，我们展示图片图片的时候通道数在后
    plt.imshow(inputs)
    # plt.show()
    plt.pause(1)
    plt.savefig(picname + '.jpg')  # 保存图片
    plt.close()


#测试
def test(modele):
    correct = 0
    total = 0
    with torch.no_grad():
        for data in test_loader:
            images, labels = data
            outputs = modele(images)
            _, predicted = torch.max(outputs.data, dim=1)
            total += labels.size(0)
            correct += (predicted == labels).sum().item()
    print('识别正确率为: %d %%' % (100 * correct / total))


#提取训练数据
def restore_net():
    net2=torch.load('cnn_1.plk')
    return net2



if __name__ == '__main__':
    '''
    for epoch in range(5):
        train(epoch)
        test(net1)
        # 保存训练结果
    torch.save(net1, 'cnn_1.plk')
    '''
    net2 = restore_net()
    test(net2)

相比较使用Linear线性层构建网络模型，卷积神经网络模型可以使正确率达到98%
运行结果