更新 function.py

function.py

@@ -30,7 +30,7 @@ name = "程悦"
 def sigmoid(z):
     return 1.0 / (1 + np.exp(-z))
 
-def train_logreg(X, y, learning_rate=0.01, iterations=1000):
+def train_logreg(X, y, learning_rate=0.01, iterations=1000, regularization='l2', lambda_=0.01):
     """
     对率回归训练函数。
 
@@ -47,6 +47,12 @@ def train_logreg(X, y, learning_rate=0.01, iterations=1000):
         z = np.dot(X, weights)
         h = sigmoid(z)
         gradient = np.dot(X.T, (h - y)) / m
+        # 正则化
+        if regularization == 'l2':
+            gradient += lambda_ * weights / m
+        elif regularization == 'l1':
+            gradient += lambda_ * np.sign(weights) / m
+
         weights -= learning_rate * gradient
     return weights
 
@@ -103,43 +109,47 @@ def load_data(filepath):
     
     return data, target_variable
 
+class CustomLogisticRegression(BaseEstimator, ClassifierMixin):
+    def __init__(self, learning_rate=0.01, iterations=1000, regularization='l2', lambda_=0.01):
+        self.learning_rate = learning_rate
+        self.iterations = iterations
+        self.regularization = regularization
+        self.lambda_ = lambda_
+        self.weights = None
+
+    def fit(self, X, y):
+        self.weights = train_logreg(X, y, self.learning_rate, self.iterations,
+                                    self.regularization, self.lambda_)
+
+    def predict(self, X):
+        return np.array([predict_logreg(x, self.weights) for x in X])
+
 if __name__ == "__main__":
     # 加载训练集和测试集
     X_train, y_train = load_data("./data/horse-colic.data")
     X_test, y_test = load_data("./data/horse-colic.test")
 
-    # 使用GridSearchCV进行超参数调优
+# 定义网格搜索的参数范围
     param_grid = {
         'learning_rate': [0.001, 0.01, 0.1],
-        'iterations': [100, 500, 1000]
+        'iterations': [100, 500, 1000],
+        'regularization': ['l1', 'l2'],
+        'lambda_': [0.01, 0.1, 1.0]
     }
 
-    # 定义自定义逻辑回归模型
+    # 创建逻辑回归模型
-    class CustomLogisticRegression:
+    model = CustomLogisticRegression()
-        def __init__(self, learning_rate=0.01, iterations=1000):
-            self.learning_rate = learning_rate
-            self.iterations = iterations
-            self.weights = None
-
-        def fit(self, X, y):
-            self.weights = train_logreg(X, y, self.learning_rate, self.iterations)
-
-        def predict(self, X):
-            return np.array([predict_logreg(x, self.weights) for x in X])
 
     # 网格搜索
-    best_accuracy = 0
+    grid_search = GridSearchCV(model, param_grid, cv=5)
-    best_params = {}
+    grid_search.fit(X_train, y_train)
-    
-    for lr in param_grid['learning_rate']:
-        for it in param_grid['iterations']:
-            model = CustomLogisticRegression(learning_rate=lr, iterations=it)
-            model.fit(X_train, y_train)
-            y_pred = model.predict(X_test)
-            accuracy = np.mean(y_pred == y_test)
-            print(f"学习率: {lr}, 迭代次数: {it}, 准确率: {accuracy}")
-            if accuracy > best_accuracy:
-                best_accuracy = accuracy
-                best_params = {'learning_rate': lr, 'iterations': it}
 
-    print(f"最佳参数: {best_params}, 最佳准确率: {best_accuracy}")
+    # 使用最佳模型进行预测
+    best_model = grid_search.best_estimator_
+    y_pred = best_model.predict(X_test)
+
+    # 测试单个数据
+    single_prediction = best_model.predict([X_test[11]])
+    print(f"最佳参数组合: {grid_search.best_params_}")
+    print(f"最佳准确率: {grid_search.best_score_}")
+    print(f"预测结果: {single_prediction[0]}, 实际结果: {y_test[11]}")