更新 function.py

function.py

@@ -35,7 +35,7 @@ def load_data(filepath):
     y = data[:, -1]
     return X, y
 
-def train_logreg(X, y, learning_rate=0.1, iterations=1000):
+def train_logreg(X, y, learning_rate=0.01, iterations=1000):
     """训练逻辑回归模型，使用随机梯度下降."""
     n_samples, n_features = X.shape
     weights = np.zeros(n_features + 1)  # 增加一位用于偏置
@@ -45,13 +45,10 @@ def train_logreg(X, y, learning_rate=0.1, iterations=1000):
         random_index = np.random.randint(n_samples)
         X_i = X[random_index]
         y_i = y[random_index]
-        
+
         linear_model = np.dot(X_i, weights)
-        prediction = 1 / (1 + np.exp(-linear_model))  # Sigmoid函数
-        error = prediction - y_i
-        
-        # 更新权重
-        weights -= learning_rate * error * X_i
+        dw = (linear_model - y_i) * X_i
+        weights -= learning_rate * dw
 
     return weights
 
@@ -59,16 +56,19 @@ def predict_logreg(inX, weights):
     """使用线性模型预测结果."""
     inX = np.append(inX, 1)  # 增加偏置项
     linear_model = np.dot(inX, weights)
-    prediction = 1 / (1 + np.exp(-linear_model))
-    return 1.0 if prediction >= 0.5 else 2.0  # 返回1.0或2.0
+    return 1.0 if linear_model >= 0 else 2.0
 
 if __name__ == "__main__":
-    # 加载训练集
+    # 示例：加载数据并训练模型
     X_train, y_train = load_data("./data/horse-colic.data")
+    weights = train_logreg(X_train, y_train, learning_rate=0.001, iterations=5000)
+
+    # 测试单个数据
+    X_test, y_test = load_data("./data/horse-colic.test")
+    y_pred = predict_logreg(X_test[11], weights)
     
-    # 数据标准化
-    scaler = StandardScaler()
-    X_train = scaler.fit_transform(X_train)
+    print(f"预测结果: {y_pred}, 实际结果: {y_test[11]}")
+
 
     # 训练逻辑回归模型
     weights = train_logreg(X_train, y_train, learning_rate=0.001, iterations=5000)