pytorch实现MNIST数据集分类

分为四个部分进行，第一步加载数据集，第二步构建模型，第三步构建损失函数和优化函数，第四步进行训练和测试。

首先导入需要用到的库

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

然后进行第一步加载数据集，需要对数据集进行转换

batch_size = 64  #设置训练的batch为64
#把训练数据转换成torch中的tensor
transform = transforms.Compose([
    transforms.ToTensor(),
    transforms.Normalize((0.1307,),(0.3081,))
])

train_dataset = datasets.MNIST(root='../dataset/mnist',
                             train=True,
                             download=False,
                             transform=transform)
train_loader = DataLoader(train_dataset,
                         shuffle=True,
                         batch_size=batch_size)
test_dataset = datasets.MNIST(root='../dataset/mnist',
                             train=False,
                             download=False,
                             transform=transform)
test_loader = DataLoader(train_dataset,
                         shuffle=False,
                         batch_size=batch_size)

第二部就是构建模型，这里构建了5层全连接网络，每一层之后再接一层非线性激活层relu，最后输出大小为10，对应了10个类别

class Net(torch.nn.Module):
    def __init__(self):
        super(Net, self).__init__()
        self.l1 = torch.nn.Linear(784, 512)
        self.l2 = torch.nn.Linear(512, 256)
        self.l3 = torch.nn.Linear(256, 128)
        self.l4 = torch.nn.Linear(128, 64)
        self.l5 = torch.nn.Linear(64, 10)
        
    def forward(self, x):
        x = x.view(-1, 784)  #将32*32的数据集展开成1*784
        x = F.relu(self.l1(x))
        x = F.relu(self.l2(x))
        x = F.relu(self.l3(x))
        x = F.relu(self.l4(x))
        return self.l5(x)
    
model = Net()

第三步就是构建损失函数和优化函数，这里使用交叉熵损失和随机梯度下降的方法进行模型优化

criterion = torch.nn.CrossEntropyLoss()
optimizer = torch.optim.SGD(model.parameters(),lr=0.01, momentum=0.5)

第四步就是模型的训练和测试

def train(epoch):
    running_loss = 0  #损失初始化
    for batch_idx, data in enumerate(train_loader,0):
        input, target = data
        optimizer.zero_grad() #将个batch的梯度清零
        output = model(input)
        loss = criterion(output, target) #计算损失
        loss.backward() #梯度反向传导
        optimizer.step() #单次优化参数更新
        
        running_loss += loss.item() 
        #没300个batch打印一次结果
        if batch_idx % 300 ==299:
            print('[%d, %5d] loss: %.3f' % (epoch + 1, batch_idx + 1, running_loss / 300))
            running_loss = 0.0

def test():
    correct = 0
    total = 0
    with torch.no_grad(): #测试过程中不用反向传导
        for data in test_loader:
            images, labels = data
            output = model(images)
            _, predicted = torch.max(output.data, dim=1)  #找到输出值最大的类别
            total += labels.size(0)
            correct += (predicted == labels).sum().item()  #计算准确率
    print('Accuracy on test set: %d %%' % (100 * correct / total))

最后运行

if __name__  == '__main__':
    for epoch in range(10):
        train(epoch)
        test()

 最后经过10轮训练，测试集精度达到了99%，效果还不错