更新 function.py

function.py

@@ -28,29 +28,55 @@ name = "邓龙树"
 import numpy as np
 from sklearn.preprocessing import StandardScaler
 
-def sigmoid(intX):  # 激活函数，逻辑回归中，将线性组合转换为概率
+def sigmoid(z):
-    return 1.0 / (1 + np.exp(-intX))
+    """Sigmoid 激活函数。"""
+    return 1.0 / (1 + np.exp(-z))
 
-def train_logreg(X, y, numIter=150):   # 训练逻辑回归模型的函数，使用梯度下降算法。
+def train_logreg(X, y, learning_rate=0.05, iterations=1500, batch_size=64, reg_lambda=0.01):
+    """训练逻辑回归模型的函数，使用梯度下降算法。"""
     m, n = X.shape
-    weights = np.zeros(n)  # 初始化权重为0而不是1
+    weights = np.random.randn(n) * 0.01  # 初始化权重
-    for j in range(numIter):
+    best_weights = weights.copy()
-        dataIndex = list(range(m))
+    best_accuracy = 0
-        for i in range(m):
+    wait = 0
-            alpha = 4 / (1.0 + j + i) + 0.01
+
-            randIndex = int(np.random.uniform(0, len(dataIndex)))
+    for i in range(iterations):
-            h = sigmoid(np.dot(X[randIndex], weights))
+        indices = np.random.permutation(m)
-            error = y[randIndex] - h
+        X_shuffled = X[indices]
-            weights = weights + alpha * error * X[randIndex]
+        y_shuffled = y[indices]
-            del(dataIndex[randIndex])
+
-    return weights
+        for j in range(0, m, batch_size):
+            X_batch = X_shuffled[j:j + batch_size]
+            y_batch = y_shuffled[j:j + batch_size]
+            z = np.dot(X_batch, weights)
+            h = sigmoid(z)
+            gradient = np.dot(X_batch.T, (h - y_batch)) / len(y_batch) + reg_lambda * weights
+
+            weights -= learning_rate * gradient
+
+        # 计算训练集准确率
+        train_accuracy = accuracy(X, y, weights)
+        if train_accuracy > best_accuracy:
+            best_accuracy = train_accuracy
+            best_weights = weights.copy()
+            wait = 0
+        else:
+            wait += 1
+
+        # 如果准确率没有提高，提前停止
+        if wait > 50:
+            break
+
+    return best_weights
 
 def predict_logreg(X, weights):
+    """预测函数。"""
     z = np.dot(X, weights)
     h = sigmoid(z)
     return np.where(h > 0.5, 1, 0)
 
 def load_data(filepath):
+    """加载数据函数。"""
     data = []
     labels = []
     with open(filepath, 'r') as f:
@@ -64,28 +90,34 @@ def load_data(filepath):
     data = np.array(data)
     labels = np.array(labels)
     target_variable = np.where((labels == 2) | (labels == 3), 0, labels)
-    data = np.delete(data, [2] + list(range(23, 28)), axis=1)
+    
+    n_features = data.shape[1]
+    columns_to_delete = [2] + list(range(23, min(28, n_features)))
+    data = np.delete(data, columns_to_delete, axis=1)
+    
     scaler = StandardScaler()
     data = scaler.fit_transform(data)
-
+    
     return data, target_variable
 
 def accuracy(X, y, weights):
+    """计算准确率。"""
     predictions = predict_logreg(X, weights)
     return np.mean(predictions == y)
 
 if __name__ == "__main__":
     X_train, y_train = load_data("./horse-colic.data")
     X_test, y_test = load_data("./horse-colic.test")
-    final_weights = train_logreg(X_train, y_train)
 
-    # 标准化数据
+    # 特征标准化
     scaler = StandardScaler()
     X_train = scaler.fit_transform(X_train)
     X_test = scaler.transform(X_test)
 
-    final_weights = train_logreg(X_train, y_train, learning_rate=0.001, iterations=1500, batch_size=64)
+    # 训练模型
+    final_weights = train_logreg(X_train, y_train, learning_rate=0.001, iterations=1500, batch_size=64, reg_lambda=0.01)
 
+    # 计算测试集准确率
     final_accuracy = accuracy(X_test, y_test, final_weights)
     print(f"最终测试集的准确率: {final_accuracy:.4f}")