更新 function.py

function.py

@@ -27,6 +27,7 @@ name = "章程鸿"
 
 # 这里写自己的代码实现
 import numpy as np
+from sklearn.model_selection import KFold
 def sigmoid(z):
     return 1 / (1 + np.exp(-z))
 
@@ -92,19 +93,12 @@ def load_data(filepath):
     y = data[:, -1]
     return X, y
 def cross_validate(X, y, k=5, learning_rate=0.01, iterations=1000):
-    m = len(y)
-    fold_size = m // k
-    indices = np.arange(m)
-    np.random.shuffle(indices)  # 打乱索引
-
+    kf = KFold(n_splits=k)
     accuracies = []
 
-    for i in range(k):
-        test_indices = indices[i * fold_size: (i + 1) * fold_size]
-        train_indices = np.concatenate((indices[:i * fold_size], indices[(i + 1) * fold_size:]))
-        
-        X_train, X_test = X[train_indices], X[test_indices]
-        y_train, y_test = y[train_indices], y[test_indices]
+    for train_index, test_index in kf.split(X):
+        X_train, X_test = X[train_index], X[test_index]
+        y_train, y_test = y[train_index], y[test_index]
 
         weights = train_logreg(X_train, y_train, learning_rate, iterations)
         predictions = [predict_logreg(x, weights) for x in X_test]
@@ -113,6 +107,7 @@ def cross_validate(X, y, k=5, learning_rate=0.01, iterations=1000):
         accuracies.append(accuracy)
 
     return np.mean(accuracies)
+
 if __name__ == "__main__":
     # 加载训练集
     X_train, y_train = load_data("./data/horse-colic.data")
@@ -120,7 +115,7 @@ if __name__ == "__main__":
     # 加载测试集
     X_test, y_test = load_data("./data/horse-colic.test")
 
-     # 进行 K 折交叉验证
+    # 进行 K 折交叉验证
     k = 5  # 设置折数
     cv_accuracy = cross_validate(X_train, y_train, k=k, learning_rate=0.01, iterations=1000)
     print(f"{k}-折交叉验证的平均准确率: {cv_accuracy:.2f}")
@@ -128,4 +123,4 @@ if __name__ == "__main__":
     # 测试单个数据
     weights = train_logreg(X_train, y_train, learning_rate=0.01, iterations=1000)
     y_pred = predict_logreg(X_test[11], weights)
-    print(f"预测结果: {y_pred}, 实际结果: {y_test[11]}")
+    print(f"预测结果: {y_pred}, 实际结果: {y_test[11]}")