实训项目使用 PyTorch 实现鸢尾花数据集

### 使用 PyTorch 实现鸢尾花数据集的实训项目 #### 数据准备 为了使用PyTorch来处理鸢尾花数据集，首先需要加载并预处理数据。可以利用`pandas`库方便地导入CSV文件，并通过`sklearn.preprocessing.StandardScaler`对特征进行标准化[^2]。 ```python import pandas as pd from sklearn.model_selection import train_test_split from sklearn.preprocessing import StandardScaler, LabelEncoder # 加载数据 url = "https://archive.ics.uci.edu/ml/machine-learning-databases/iris/iris.data" column_names = ['sepal_length', 'sepal_width', 'petal_length', 'petal_width', 'species'] data = pd.read_csv(url, names=column_names) # 编码标签 label_encoder = LabelEncoder() data['species'] = label_encoder.fit_transform(data['species']) # 划分训练集和测试集 X_train, X_test, y_train, y_test = train_test_split( data.drop('species', axis=1), data['species'], test_size=0.2, random_state=42 ) # 特征缩放 scaler = StandardScaler().fit(X_train) X_train_scaled = scaler.transform(X_train) X_test_scaled = scaler.transform(X_test) ``` #### 创建神经网络模型 定义一个多层感知器（MLP），这是一个简单的全连接前馈神经网络结构，在这里采用两层隐藏层的设计方案： ```python import torch import torch.nn as nn import torch.optim as optim class IrisNet(nn.Module): def __init__(self): super(IrisNet, self).__init__() self.fc1 = nn.Linear(4, 16) # 输入维度为4，输出维度为16 self.fc2 = nn.Linear(16, 8) # 隐藏层之间的转换 self.fc3 = nn.Linear(8, 3) # 输出类别数为3 def forward(self, x): x = torch.relu(self.fc1(x)) x = torch.relu(self.fc2(x)) x = self.fc3(x) return x model = IrisNet() criterion = nn.CrossEntropyLoss() # 定义损失函数 optimizer = optim.Adam(model.parameters(), lr=0.01) # Adam优化算法 ``` #### 训练过程 将上述准备工作完成后就可以进入正式的训练环节了。需要注意的是每次迭代都要清空梯度缓存以防止累积错误影响后续更新操作的结果准确性。 ```python num_epochs = 500 for epoch in range(num_epochs): model.train() inputs = torch.tensor(X_train_scaled, dtype=torch.float32) labels = torch.tensor(y_train.values, dtype=torch.long) optimizer.zero_grad() # 清除之前的梯度信息 outputs = model(inputs) loss = criterion(outputs, labels) loss.backward() # 反向传播计算参数梯度 optimizer.step() # 更新权重参数 if (epoch+1) % 100 == 0: print(f'Epoch [{epoch+1}/{num_epochs}], Loss: {loss.item():.4f}') ``` #### 测试评估 完成训练之后还需要验证所建立起来的分类器性能如何，通常可以通过准确率等指标来进行衡量。 ```python with torch.no_grad(): correct = 0 total = 0 for i in range(len(X_test)): input_tensor = torch.tensor([X_test_scaled[i]], dtype=torch.float32) output = model(input_tensor) _, predicted = torch.max(output.data, 1) if predicted.item() == y_test.iloc[i]: correct += 1 total += 1 accuracy = correct / total * 100 print(f'Test Accuracy of the model on the test set is: {accuracy:.2f}%') ```