單通道圖片的規范寫法
import torch
import torch.nn as nn
import torch.optim as optim
from torch.utils.data import DataLoader , Dataset
from torchvision import datasets, transforms
import matplotlib.pyplot as plt
import warnings
warnings.filterwarnings("ignore")
torch.manual_seed(42)
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
print(f"使用設備: {device}")transform = transforms.Compose([transforms.ToTensor(), transforms.Normalize((0.1307,), (0.3081,))
])train_dataset = datasets.MNIST(root='./data',train=True,download=True,transform=transform
)test_dataset = datasets.MNIST(root='./data',train=False,transform=transform
)batch_size = 64
train_loader = DataLoader(train_dataset, batch_size=batch_size, shuffle=True)
test_loader = DataLoader(test_dataset, batch_size=batch_size, shuffle=False)class MLP(nn.Module):def __init__(self):super(MLP, self).__init__()self.flatten = nn.Flatten() self.layer1 = nn.Linear(784, 128) self.relu = nn.ReLU() # 激活函數self.layer2 = nn.Linear(128, 10) def forward(self, x):x = self.flatten(x) x = self.layer1(x) x = self.relu(x) x = self.layer2(x) return xmodel = MLP()
model = model.to(device) criterion = nn.CrossEntropyLoss()
optimizer = optim.Adam(model.parameters(), lr=0.001) def train(model, train_loader, test_loader, criterion, optimizer, device, epochs):model.train()all_iter_losses = [] iter_indices = [] for epoch in range(epochs):running_loss = 0.0correct = 0total = 0for batch_idx, (data, target) in enumerate(train_loader):data, target = data.to(device), target.to(device) # 移至GPU(如果可用)optimizer.zero_grad() output = model(data) loss = criterion(output, target) loss.backward() optimizer.step() iter_loss = loss.item()all_iter_losses.append(iter_loss)iter_indices.append(epoch * len(train_loader) + batch_idx + 1)running_loss += loss.item() _, predicted = output.max(1) total += target.size(0) correct += predicted.eq(target).sum().item() if (batch_idx + 1) % 100 == 0:print(f'Epoch: {epoch+1}/{epochs} | Batch: {batch_idx+1}/{len(train_loader)} 'f'| 單Batch損失: {iter_loss:.4f} | 累計平均損失: {running_loss/(batch_idx+1):.4f}')epoch_train_loss = running_loss / len(train_loader)epoch_train_acc = 100. * correct / totalepoch_test_loss, epoch_test_acc = test(model, test_loader, criterion, device)print(f'Epoch {epoch+1}/{epochs} 完成 | 訓練準確率: {epoch_train_acc:.2f}% | 測試準確率: {epoch_test_acc:.2f}%')plot_iter_losses(all_iter_losses, iter_indices)return epoch_test_acc
測試函數和繪圖函數均被封裝在了train函數中，但是test和繪圖函數在定義train函數之后，這是因為在 Python 中，函數定義的順序不影響調用，只要在調用前已經完成定義即可。
#測試模型（不變）
def test(model, test_loader, criterion, device):model.eval() # 設置為評估模式test_loss = 0correct = 0total = 0with torch.no_grad(): # 不計算梯度，節省內存和計算資源for data, target in test_loader:data, target = data.to(device), target.to(device)output = model(data)test_loss += criterion(output, target).item()_, predicted = output.max(1)total += target.size(0)correct += predicted.eq(target).sum().item()avg_loss = test_loss / len(test_loader)accuracy = 100. * correct / totalreturn avg_loss, accuracy # 返回損失和準確率# 繪制每個 iteration 的損失曲線
def plot_iter_losses(losses, indices):plt.figure(figsize=(10, 4))plt.plot(indices, losses, 'b-', alpha=0.7, label='Iteration Loss')plt.xlabel('Iteration（Batch序號）')plt.ylabel('損失值')plt.title('每個 Iteration 的訓練損失')plt.legend()plt.grid(True)plt.tight_layout()plt.show()# 執行訓練和測試（設置 epochs=2 驗證效果）
epochs = 2
print("開始訓練模型...")
final_accuracy = train(model, train_loader, test_loader, criterion, optimizer, device, epochs)
print(f"訓練完成！最終測試準確率: {final_accuracy:.2f}%")下面是彩色圖片的規范寫法 ，彩色通道也是在第一步被直接展平，其他代碼一致
import torch
import torch.nn as nn
import torch.optim as optim
from torchvision import datasets, transforms
from torch.utils.data import DataLoader
import matplotlib.pyplot as plt
import numpy as np# 設置中文字體支持
plt.rcParams["font.family"] = ["SimHei"]
plt.rcParams['axes.unicode_minus'] = False # 解決負號顯示問題# 1. 數據預處理
transform = transforms.Compose([transforms.ToTensor(), # 轉換為張量transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5)) # 標準化處理
])# 2. 加載CIFAR-10數據集
train_dataset = datasets.CIFAR10(root='./data',train=True,download=True,transform=transform
)test_dataset = datasets.CIFAR10(root='./data',train=False,transform=transform
)# 3. 創建數據加載器
batch_size = 64
train_loader = DataLoader(train_dataset, batch_size=batch_size, shuffle=True)
test_loader = DataLoader(test_dataset, batch_size=batch_size, shuffle=False)# 4. 定義MLP模型（適應CIFAR-10的輸入尺寸）
class MLP(nn.Module):def __init__(self):super(MLP, self).__init__()self.flatten = nn.Flatten() # 將3x32x32的圖像展平為3072維向量self.layer1 = nn.Linear(3072, 512) # 第一層：3072個輸入，512個神經元self.relu1 = nn.ReLU()self.dropout1 = nn.Dropout(0.2) # 添加Dropout防止過擬合self.layer2 = nn.Linear(512, 256) # 第二層：512個輸入，256個神經元self.relu2 = nn.ReLU()self.dropout2 = nn.Dropout(0.2)self.layer3 = nn.Linear(256, 10) # 輸出層：10個類別def forward(self, x):# 第一步：將輸入圖像展平為一維向量x = self.flatten(x) # 輸入尺寸: [batch_size, 3, 32, 32] → [batch_size, 3072]# 第一層全連接 + 激活 + Dropoutx = self.layer1(x) # 線性變換: [batch_size, 3072] → [batch_size, 512]x = self.relu1(x) # 應用ReLU激活函數x = self.dropout1(x) # 訓練時隨機丟棄部分神經元輸出# 第二層全連接 + 激活 + Dropoutx = self.layer2(x) # 線性變換: [batch_size, 512] → [batch_size, 256]x = self.relu2(x) # 應用ReLU激活函數x = self.dropout2(x) # 訓練時隨機丟棄部分神經元輸出# 第三層（輸出層）全連接x = self.layer3(x) # 線性變換: [batch_size, 256] → [batch_size, 10]return x # 返回未經過Softmax的logits# 檢查GPU是否可用
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")# 初始化模型
model = MLP()
model = model.to(device) # 將模型移至GPU（如果可用）criterion = nn.CrossEntropyLoss() # 交叉熵損失函數
optimizer = optim.Adam(model.parameters(), lr=0.001) # Adam優化器# 5. 訓練模型（記錄每個 iteration 的損失）
def train(model, train_loader, test_loader, criterion, optimizer, device, epochs):model.train() # 設置為訓練模式# 記錄每個 iteration 的損失all_iter_losses = [] # 存儲所有 batch 的損失iter_indices = [] # 存儲 iteration 序號for epoch in range(epochs):running_loss = 0.0correct = 0total = 0for batch_idx, (data, target) in enumerate(train_loader):data, target = data.to(device), target.to(device) # 移至GPUoptimizer.zero_grad() # 梯度清零output = model(data) # 前向傳播loss = criterion(output, target) # 計算損失loss.backward() # 反向傳播optimizer.step() # 更新參數# 記錄當前 iteration 的損失iter_loss = loss.item()all_iter_losses.append(iter_loss)iter_indices.append(epoch * len(train_loader) + batch_idx + 1)# 統計準確率和損失running_loss += iter_loss_, predicted = output.max(1)total += target.size(0)correct += predicted.eq(target).sum().item()# 每100個批次打印一次訓練信息if (batch_idx + 1) % 100 == 0:print(f'Epoch: {epoch+1}/{epochs} | Batch: {batch_idx+1}/{len(train_loader)} 'f'| 單Batch損失: {iter_loss:.4f} | 累計平均損失: {running_loss/(batch_idx+1):.4f}')# 計算當前epoch的平均訓練損失和準確率epoch_train_loss = running_loss / len(train_loader)epoch_train_acc = 100. * correct / total# 測試階段model.eval() # 設置為評估模式test_loss = 0correct_test = 0total_test = 0with torch.no_grad():for data, target in test_loader:data, target = data.to(device), target.to(device)output = model(data)test_loss += criterion(output, target).item()_, predicted = output.max(1)total_test += target.size(0)correct_test += predicted.eq(target).sum().item()epoch_test_loss = test_loss / len(test_loader)epoch_test_acc = 100. * correct_test / total_testprint(f'Epoch {epoch+1}/{epochs} 完成 | 訓練準確率: {epoch_train_acc:.2f}% | 測試準確率: {epoch_test_acc:.2f}%')# 繪制所有 iteration 的損失曲線plot_iter_losses(all_iter_losses, iter_indices)return epoch_test_acc # 返回最終測試準確率# 6. 繪制每個 iteration 的損失曲線
def plot_iter_losses(losses, indices):plt.figure(figsize=(10, 4))plt.plot(indices, losses, 'b-', alpha=0.7, label='Iteration Loss')plt.xlabel('Iteration（Batch序號）')plt.ylabel('損失值')plt.title('每個 Iteration 的訓練損失')plt.legend()plt.grid(True)plt.tight_layout()plt.show()# 7. 執行訓練和測試
epochs = 20 # 增加訓練輪次以獲得更好效果
print("開始訓練模型...")
final_accuracy = train(model, train_loader, test_loader, criterion, optimizer, device, epochs)
print(f"訓練完成！最終測試準確率: {final_accuracy:.2f}%")# # 保存模型
# torch.save(model.state_dict(), 'cifar10_mlp_model.pth')
# # print("模型已保存為: cifar10_mlp_model.pth")
