更新 function.py

function.py

@@ -34,45 +34,80 @@ def sigmoid(z):
     """Sigmoid 激活函数。"""
     return 1 / (1 + np.exp(-z))
 
-def train_logreg(X, y, learning_rate=0.1, iterations=2000, lambda_reg=0.01, batch_size=32):
+def train_logreg(X, y, other_params=None):
-    """逻辑回归训练函数，随机初始化权重，并在每次迭代前打乱数据。"""
+    """
+    对率回归训练函数。
+
+    Args:
+    - X (numpy array): 特征数据。
+    - y (numpy array): 标签数据。
+    - other_params (dict): 其它参数，可选设置。
+
+    Returns:
+    - numpy array: 训练得到的权重。
+    """
+    if other_params is None:
+        other_params = {
+            'learning_rate': 0.1,
+            'iterations': 2000,
+            'lambda_reg': 0.01,
+            'batch_size': 32
+        }
+    
+    learning_rate = other_params.get('learning_rate', 0.1)
+    iterations = other_params.get('iterations', 2000)
+    lambda_reg = other_params.get('lambda_reg', 0.01)
+    batch_size = other_params.get('batch_size', 32)
+
     m, n = X.shape
     weights = np.random.randn(n) * 0.01  # 随机初始化权重，取较小值
 
     for i in range(iterations):
-        # 每次迭代打乱数据
         indices = np.random.permutation(m)
-        X = X[indices]
+        X_shuffled = X[indices]
-        y = y[indices]
+        y_shuffled = y[indices]
 
-        for j in range(0, m, batch_size):  # 按照 batch size 迭代
+        for j in range(0, m, batch_size):
-            X_batch = X[j:j + batch_size]
+            X_batch = X_shuffled[j:j + batch_size]
-            y_batch = y[j:j + batch_size]
+            y_batch = y_shuffled[j:j + batch_size]
 
             z = np.dot(X_batch, weights)
             predictions = sigmoid(z)
-            gradient = np.dot(X_batch.T, (predictions - y_batch)) / X_batch.shape[0]
+            gradient = np.dot(X_batch.T, (predictions - y_batch)) / X_batch.shape[0] + (lambda_reg / m) * weights
 
-            # L2正则化
-            gradient += (lambda_reg / m) * weights
-
-            # 更新权重
             weights -= learning_rate * gradient
 
-        # 动态调整学习率，逐渐减小
+        if i % 100 == 0:
-        if i % 100 == 0:  # 每100次迭代输出一次学习率
             print(f"Iteration {i}, Learning Rate: {learning_rate}")
 
     return weights
 
 def predict_logreg(inX, weights):
-    """逻辑回归预测函数。"""
+    """
+    对率回归预测函数。
+
+    Args:
+    - inX (numpy array): 输入待分类的样本。
+    - weights (numpy array): 训练得到的权重。
+
+    Returns:
+    - int: 预测的分类标签。
+    """
     z = np.dot(inX, weights)
     prediction_prob = sigmoid(z)
     return 1 if prediction_prob >= 0.5 else 0
 
 def load_data(filepath):
-    """加载数据并处理缺失值。"""
+    """
+    从文件中加载数据。
+
+    Args:
+    - filepath (str): 数据文件路径。
+
+    Returns:
+    - np.array: 特征数据。
+    - np.array: 标签数据。
+    """
     data = []
     labels = []
 
@@ -81,18 +116,15 @@ def load_data(filepath):
             values = line.strip().replace('?', '0').split()
             try:
                 values = [float(v) for v in values]
-                
+                if values[22] == 2 or values[22] == 3:
-                # 合并特征23的取值
+                    labels.append(0)
-                if values[22] == 2 or values[22] == 3:  # 特征23（索引22）
-                    labels.append(0)  # 死亡
                 elif values[22] == 1:
-                    labels.append(1)  # 生存
+                    labels.append(1)
                 else:
-                    continue  # 无效标签，丢弃
+                    continue
                 
-                # 用0填充缺失值
+                values = [0 if v == 0 else v for v in values]
-                values = [0 if v == 0 else v for v in values]  # 直接用0填充
+                data.append(values[:2] + values[3:23])
-                data.append(values[:2] + values[3:23])  # 保留特征1, 2 和特征4-23
             except ValueError:
                 continue
 
@@ -101,44 +133,22 @@ def load_data(filepath):
 
     return data, labels
 
-
-def preprocess_data(X):
-    """特征标准化。"""
-    scaler = StandardScaler()
-    return scaler.fit_transform(X)
-
 if __name__ == "__main__":
-    # 加载数据集
+    # 加载训练集
-    X, y = load_data("./data/horse-colic.data")
+    X_train, y_train = load_data("./data/horse-colic.data")
 
-    # 归一化数据
+    # 加载测试集
-    X = preprocess_data(X)
+    X_test, y_test = load_data("./data/horse-colic.test")
 
-    # 使用 train_test_split 打乱并随机抽取数据
+    # 训练对率回归模型
-    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
+    other_params = {
-
+        'learning_rate': 0.01,
-    # K折交叉验证
+        'iterations': 1000,
-    k = 5
+        'lambda_reg': 0.01,
-    kf = KFold(n_splits=k, shuffle=True, random_state=42)
+        'batch_size': 32
-    accuracies = []
+    }
-
+    weights = train_logreg(X_train, y_train, other_params=other_params)
-    for train_index, val_index in kf.split(X_train):
-        X_train_kf, X_val_kf = X_train[train_index], X_train[val_index]
-        y_train_kf, y_val_kf = y_train[train_index], y_train[val_index]
-
-        # 训练逻辑回归模型，设置 batch size
-        batch_size = 32
-        weights = train_logreg(X_train_kf, y_train_kf, learning_rate=0.1, iterations=2000, lambda_reg=0.01, batch_size=batch_size)
-
-        # 验证模型
-        val_predictions = [predict_logreg(X_val_kf[i], weights) for i in range(len(X_val_kf))]
-        accuracy = np.mean(val_predictions == y_val_kf)
-        accuracies.append(accuracy)
-
-    # 输出交叉验证的准确率和平均准确率
-    print(f"K折交叉验证得分: {accuracies}")
-    print(f"平均准确率: {np.mean(accuracies)}")
 
     # 测试单个数据
     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]}")