Pytorch兩種構建網絡的方法

版本：Pytorch 1.0  代碼是在jupter中執行的。

導包：

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

設置超參：

BATCH_SIZE = 512 # 大概需要2G的顯存
EPOCHS = 20 # 總共訓練批次
DEVICE = torch.device("cuda" if torch.cuda.is_available() else "cpu") 

加載數據：

# 下載訓練集
train_loader = torch.utils.data.DataLoader(
    datasets.MNIST('data', train = True, download = True,
              transform = transforms.Compose([
                  transforms.ToTensor(),
                  transforms.Normalize((0.1037,), (0.3081,))
              ])),
batch_size = BATCH_SIZE, shuffle = True)

# 測試集
test_loader = torch.utils.data.DataLoader(
datasets.MNIST('data', train = False, transform = transforms.Compose([
    transforms.ToTensor(),
    transforms.Normalize((0.1037,), (0.3081,))
])),
batch_size = BATCH_SIZE, shuffle = True)

構建網絡：方式一

# 定義模型
class ConvNet(nn.Module):
    def __init__(self):
        super().__init__()
        #1*1*28*28
        self.conv1 = nn.Conv2d(1, 10, 5) 
        self.conv2 = nn.Conv2d(10, 20, 3) 
        self.fc1 = nn.Linear(20 * 10 * 10, 500)
        self.fc2 = nn.Linear(500, 10)
        
    def forward(self, x):
        in_size = x.size(0)
        out= self.conv1(x) # 1* 10 * 24 *24
        out = F.relu(out)
        out = F.max_pool2d(out, 2, 2) # 1* 10 * 12 * 12
        out = self.conv2(out) # 1* 20 * 10 * 10
        out = F.relu(out)
        out = out.view(in_size, -1) # 1 * 2000
        out = self.fc1(out) # 1 * 500
        out = F.relu(out)
        out = self.fc2(out) # 1 * 10
        out = F.log_softmax(out, dim = 1)
        return out

構建網絡：方式二——把更多的內容放在了Sequential里面，覺得網絡會顯得清楚一些

class MyNet(torch.nn.Module):
    def __init__(self):  
        super(MyNet, self).__init__()  
        self.conv1 = nn.Sequential( # (1,28,28)  
                     nn.Conv2d(in_channels=1, out_channels=16, kernel_size=5,  
                               stride=1, padding=2), # (16,28,28)  
        # 想要con2d卷積出來的圖片尺寸沒有變化, padding=(kernel_size-1)/2  
                     nn.ReLU(),  
                     nn.MaxPool2d(kernel_size=2) # (16,14,14)  
                     )  
        self.conv2 = nn.Sequential( # (16,14,14)  
                     nn.Conv2d(16, 32, 5, 1, 2), # (32,14,14)  
                     nn.ReLU(),  
                     nn.MaxPool2d(2) # (32,7,7)  
                     )  
        self.out = nn.Linear(32*7*7, 10)  
  
    def forward(self, x):  
        x = self.conv1(x)  
        x = self.conv2(x)  
        x = x.view(x.size(0), -1) # 將（batch，32,7,7）展平為（batch，32*7*7）  
        output = self.out(x)  
        return output 

定義優化器：

#生成模型和優化器
model = MyNet().to(DEVICE)  # MyNet可以改為ConvNet 調用不同模型
optimizer = optim.Adam(model.parameters())

定義訓練和測試函數：

# 定義訓練函數
def train(model, device, train_loader, optimizer, epoch):
    model.train()  # 設置為trainning模式
    for batch_idx, (data, target) in enumerate(train_loader):
        data, target = data.to(device), target.to(device)
        optimizer.zero_grad()  # 優化器梯度初始化為零
        output = model(data)  # 把數據輸入網絡並得到輸出，即進行前向傳播
        loss = F.cross_entropy(output, target)   # 定義損失函數
        loss.backward()  # 反向傳播梯度
        optimizer.step()  # 結束一次前傳+反傳之后，更新參數
        if (batch_idx + 1) % 30 == 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()))
# 定義測試函數
def test(model, device, test_loader):
    model.eval()  # 設置為test模式
    test_loss =0  # 初始化測試損失值為0
    correct = 0  # 初始化預測正確的數據個數為0
    with torch.no_grad():
        for data, target in test_loader:
            data, target = data.to(device), target.to(device)  #計算前要把變量變成Variable形式，因為這樣子才有梯度
            output = model(data)
            test_loss += F.nll_loss(output, target, reduction = 'sum') # 將一批的損失相加
            pred = output.max(1, keepdim = True)[1] # 找到概率最大的下標
            correct += pred.eq(target.view_as(pred)).sum().item()  # 對預測正確的數據個數進行累加
    
    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)
            ))

main函數

# 最后開始訓練和測試
for epoch in range(1, EPOCHS + 1):
    train(model,  DEVICE, train_loader, optimizer, epoch)
    test(model, DEVICE, test_loader)