更新 function.py

function.py

@@ -1,28 +1,27 @@
 import numpy as np
 import os
-from sklearn.preprocessing import StandardScaler  # Import standardization library
 
-# Student ID and Name
+# 学号和姓名
 id = "202210423220"
 name = "方子涵"
 
 def sigmoid(z):
-    """Calculate the sigmoid function"""
+    """计算sigmoid函数"""
     return 1.0 / (1 + np.exp(-z))
 
 def compute_gradient(X_batch, y_batch, weights, lambda_reg):
-    """Calculate the gradient, including L2 regularization"""
+    """计算梯度，包含L2正则化"""
     z = np.dot(X_batch, weights)
     h = sigmoid(z)
     gradient = np.dot(X_batch.T, (h - y_batch)) / len(y_batch)
-    gradient += (lambda_reg / len(y_batch)) * weights  # L2 regularization
+    gradient += (lambda_reg / len(y_batch)) * weights  # L2正则化
     return gradient
 
 def train_logreg(X, y, learning_rate=0.001, iterations=10000, batch_size=64, lambda_reg=0.1):
-    """Logistic regression training function."""
+    """对率回归训练函数。"""
     m, n = X.shape
-    weights = np.random.randn(n) * np.sqrt(2. / n)  # He initialization
-    print("Starting to train the logistic regression model")
+    weights = np.random.randn(n) * np.sqrt(2. / n)  # He初始化
+    print("开始训练对率回归模型")
     
     for i in range(iterations):
         indices = np.random.permutation(m)
@@ -35,21 +34,27 @@ def train_logreg(X, y, learning_rate=0.001, iterations=10000, batch_size=64, lam
             gradient = compute_gradient(X_batch, y_batch, weights, lambda_reg)
             weights -= learning_rate * gradient
             
-        if i % 1000 == 0:  # Print information every 1000 iterations
-            print(f"Iteration {i}: Current weight shape: {weights.shape}")
+        if i % 1000 == 0:  # 每1000次迭代打印一次信息
+            print(f"迭代 {i}: 当前权重形状: {weights.shape}")
 
     return weights
 
 def predict_logreg(inX, weights):
-    """Logistic regression prediction function."""
+    """对率回归预测函数。"""
     z = np.dot(inX, weights)
     h = sigmoid(z)
     return 1 if h >= 0.5 else 0
 
+def standardize_data(X):
+    """标准化数据"""
+    mean = np.mean(X, axis=0)
+    std = np.std(X, axis=0)
+    return (X - mean) / std, mean, std
+
 def load_data(filepath):
-    """Load data from a file."""
+    """从文件中加载数据。"""
     if not os.path.exists(filepath):
-        raise FileNotFoundError(f"File {filepath} does not exist!")
+        raise FileNotFoundError(f"文件 {filepath} 不存在！")
     
     data, labels = [], []
     print(f"Loading data: {filepath}")
@@ -65,7 +70,7 @@ def load_data(filepath):
     labels = np.array(labels)
     
     num_features = data.shape[1]
-    indices_to_remove = [2]  # Remove feature 3 (index 2)
+    indices_to_remove = [2]  # 去掉特征3（索引2）
     
     if num_features > 23:
         indices_to_remove.extend(range(23, min(28, num_features)))
@@ -77,7 +82,7 @@ def load_data(filepath):
     return data, target_variable
 
 def k_fold_cross_validation(X, y, k=5, learning_rate=0.001, iterations=10000, batch_size=64, lambda_reg=0.1):
-    """K-fold cross-validation."""
+    """K折交叉验证。"""
     m = X.shape[0]
     fold_size = m // k
     accuracies = []
@@ -101,34 +106,33 @@ def k_fold_cross_validation(X, y, k=5, learning_rate=0.001, iterations=10000, ba
     return np.mean(accuracies)
 
 if __name__ == "__main__":
-    # Load training set
+    # 加载训练集
     try:
         X_train, y_train = load_data("./data/horse-colic.data")
     except FileNotFoundError as e:
         print(e)
         exit()
 
-    # Data normalization
-    scaler = StandardScaler()
-    X_train = scaler.fit_transform(X_train)  # Standardize using the training set
+    # 数据标准化处理
+    X_train, mean, std = standardize_data(X_train)
 
-    # K-fold cross-validation
+    # K折交叉验证
     avg_accuracy = k_fold_cross_validation(X_train, y_train, k=5, learning_rate=0.01, iterations=1000)
     print(f"Average accuracy of K-fold cross-validation: {avg_accuracy:.4f}")
 
-    # Load test set
+    # 加载测试集
     try:
         X_test, y_test = load_data("./data/horse-colic.test")
     except FileNotFoundError as e:
         print(e)
         exit()
 
-    # Test set also needs the same normalization
-    X_test = scaler.transform(X_test)  # Standardize using the same Scaler
+    # 测试集也需要进行相同的标准化
+    X_test = (X_test - mean) / std  # 使用训练集的均值和标准差进行标准化
 
-    # Train logistic regression model
+    # 训练对率回归模型
     weights = train_logreg(X_train, y_train, learning_rate=0.01, iterations=1000)
 
-    # Test a single data point
+    # 测试单个数据
     y_pred = predict_logreg(X_test[11], weights)
     print(f"Predicted result: {y_pred}, Actual result: {y_test[11]}")