pytorch学习笔记（十二）————全连接层

nn.Linear()

class torch.nn.Linear（in_features，out_features，bias = True ）

对传入数据应用线性变换：y = A x+ b

参数：

in_features - 每个输入样本的大小

out_features - 每个输出样本的大小

bias - 如果设置为False，则图层不会学习附加偏差。默认值：True

实现代码：

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


batch_size=200
learning_rate=0.01
epochs=10

train_loader = torch.utils.data.DataLoader(
    datasets.MNIST('../data', train=True, download=True,
                   transform=transforms.Compose([
                       transforms.ToTensor(),
                       transforms.Normalize((0.1307,), (0.3081,))
                   ])),
    batch_size=batch_size, shuffle=True)
test_loader = torch.utils.data.DataLoader(
    datasets.MNIST('../data', train=False, transform=transforms.Compose([
        transforms.ToTensor(),
        transforms.Normalize((0.1307,), (0.3081,))
    ])),
    batch_size=batch_size, shuffle=True)



class MLP(nn.Module):#自定义类 继承nn.Module

    def __init__(self):#初始化函数
        super(MLP, self).__init__()#继承父类初始化函数

        self.model = nn.Sequential(
            nn.Linear(784, 200),
            nn.ReLU(inplace=True),
            nn.Linear(200, 200),
            nn.ReLU(inplace=True),
            nn.Linear(200, 10),
            nn.ReLU(inplace=True),
        )#自定义实例属性 model 传入自定义模型的内部构造 返回类

    def forward(self, x):
        x = self.model(x)
        #x传入自定义的model类 返回经过模型后的输出
        return x

net = MLP()#创建实例
optimizer = optim.SGD(net.parameters(), lr=learning_rate)
#优化实例的所有对象
criteon = nn.CrossEntropyLoss()
#创建交叉熵实例
for epoch in range(epochs):

    for batch_idx, (data, target) in enumerate(train_loader):
        data = data.view(-1, 28*28)

        logits = net(data)#得到x经过模型后的输出
        loss = criteon(logits, target)
        #得到loss
        optimizer.zero_grad()
        loss.backward()
        # print(w1.grad.norm(), w2.grad.norm())
        optimizer.step()

        if batch_idx % 100 == 0:
            print('Train Epoch: {} [{}/{} ({:.0f}%)]\tLoss: {:.6f}'.format(
                epoch, batch_idx * len(data), len(train_loader.dataset),
                       100. * batch_idx / len(train_loader), loss.item()))


    test_loss = 0
    correct = 0
    for data, target in test_loader:
        data = data.view(-1, 28 * 28)
        logits = net(data)
        test_loss += criteon(logits, target).item()

        pred = logits.data.max(1)[1]
        correct += pred.eq(target.data).sum()

    test_loss /= len(test_loader.dataset)
    print('\nTest set: Average loss: {:.4f}, Accuracy: {}/{} ({:.0f}%)\n'.format(
        test_loss, correct, len(test_loader.dataset),
        100. * correct / len(test_loader.dataset)))