更新 function.py

function.py

@@ -1,6 +1,3 @@
-import numpy as np
-import pandas as pd
-
 """
 实验说明:
 - 自行编写实现数据加载与预测病马死亡的对率回归算法。
@@ -19,92 +16,115 @@ import pandas as pd
 
 """
 请在这里填写学号和姓名，不填写即成绩无效。
-姓名请直接**填写中文**
-例如: name = "张三"
 """
 id = "202210423115"
 name = "李煜锋"
 
-def load_data(filepath):
-    """
-    从文件中加载数据，若不自行正确实现，则无法通过评测，**不记录成绩**。
+import numpy as np
+from sklearn.preprocessing import MinMaxScaler
+from sklearn.model_selection import train_test_split
 
-    Args:
-    - filepath (str): 数据文件路径。
+def sigmoid(z):
+    return 1.0 / (1 + np.exp(-z))
 
-    Returns:
-    - np.array: 特征数据。
-    - np.array: 标签数据。
-    """
-    # 使用空白作为分隔符读取数据
-    data = pd.read_csv(filepath, sep='\s+', header=None, na_values='?')
+def compute_gradient(X_batch, y_batch, weights, lambda_reg):
+    z = np.dot(X_batch, weights)
+    h = sigmoid(z)
+    gradient = np.dot(X_batch.T, (h - y_batch)) / len(y_batch)
+    gradient += (lambda_reg / len(y_batch)) * weights  # L2正则化
+    return gradient
 
-    # 填充缺失值
-    data.fillna(0, inplace=True)
-
-    # 处理目标变量，假设第 23 列是目标变量（索引 22）
-    data[22] = data[22].replace({2: 0, 3: 0})
-
-    # 去掉特征 3（索引 2）及特征 24~28（索引 23~27），如果这些列存在
-    columns_to_drop = [2, 23, 24, 25, 26, 27]
-    data = data.drop(columns=[col for col in columns_to_drop if col in data.columns])
-
-    # 分离特征和标签
-    X = data.drop(columns=[22]).to_numpy()  # 特征数据
-    y = data[22].to_numpy()  # 标签数据
-    return X, y
-
-def train_logreg(X, y, learning_rate=0.01, iterations=1000):
-    """
-    对率回归训练函数。
-
-    Args:
-    - X (numpy array): 特征数据。
-    - y (numpy array): 标签数据。
-    - learning_rate (float): 学习率，默认值为0.01。
-    - iterations (int): 迭代次数，默认值为1000。
-
-    Returns:
-    - numpy array: 训练得到的权重。
-    """
+def train_logreg(X, y, learning_rate=0.001, iterations=10000, batch_size=64, lambda_reg=0.1):
     m, n = X.shape
-    weights = np.zeros(n)
+    weights = np.random.randn(n) * np.sqrt(2. / n)  # He初始化
+    best_weights = weights.copy()
+    best_accuracy = 0
 
-    # 梯度下降法
     for _ in range(iterations):
-        linear_model = np.dot(X, weights)
-        predictions = 1 / (1 + np.exp(-linear_model))  # sigmoid函数
-        # 更新权重
-        weights += learning_rate * (1/m) * np.dot(X.T, (y - predictions))
+        indices = np.random.permutation(m)
+        X_shuffled = X[indices]
+        y_shuffled = y[indices]
 
-    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]
+            gradient = compute_gradient(X_batch, y_batch, weights, lambda_reg)
+            weights -= learning_rate * gradient
 
-def predict_logreg(inX, weights):
-    """
-    对率回归预测函数。
+        # 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()
 
-    Args:
-    - inX (numpy array): 输入待分类的样本。
-    - weights (numpy array): 训练得到的权重。
+        # 动态调整学习率
+        if _ % 100 == 0 and train_accuracy > 0.95:  # 达到高准确率时减小学习率
+            learning_rate *= 0.9
 
-    Returns:
-    - int: 预测的分类标签。
-    """
-    linear_model = np.dot(inX, weights)
-    prediction = 1 / (1 + np.exp(-linear_model))  # sigmoid函数
-    return 1 if prediction >= 0.5 else 0  # 采用0.5作为阈值
+    return best_weights
 
+def predict_logreg(X, weights):
+    z = np.dot(X, weights)
+    h = sigmoid(z)
+    return np.where(h >= 0.5, 1, 0)
+
+def load_data(filepath):
+    data, labels = [], []
+    with open(filepath, 'r') as f:
+        for line in f:
+            line = line.strip().split()
+            if len(line) >= 1 and line[-1].isdigit():
+                line_data = [0 if x == '?' else float(x) for x in line[:-1]]
+                data.append(line_data)
+                labels.append(int(line[-1]))
+
+    data = np.array(data)
+    labels = np.array(labels)
+
+    # 检查特征数量
+    num_features = data.shape[1]
+
+    # 特征选择：去掉特征3（索引2）和特征24~28（索引23~26），确保索引有效
+    indices_to_delete = [2]  # 特征3
+    if num_features > 24:  # 确保特征数量足够
+        indices_to_delete += list(range(24, 27))  # 特征24到27（索引23到26）
+
+    # 删除特征
+    data = np.delete(data, indices_to_delete, axis=1)
+
+    # 将标签二分类化，特征23（索引22）取值2和3合并为0，其余为1
+    target_variable = np.where((labels == 2) | (labels == 3), 0, 1)
+
+    return data, target_variable
+
+def accuracy(X, y, weights):
+    predictions = predict_logreg(X, weights)
+    return np.mean(predictions == y)
+
+# 其余代码保持不变
 if __name__ == "__main__":
-    # 以下代码可用于测试，可自行随意修改
-    # 加载训练集
-    X_train, y_train = load_data("./horse-colic.data")
+    # 加载数据
+    X, y = load_data("./data/horse-colic.data")
+
+    # 拆分训练集和验证集
+    X_train, X_val, y_train, y_val = train_test_split(X, y, test_size=0.2, random_state=42)
+
+    # 数据归一化
+    scaler = MinMaxScaler()
+    X_train = scaler.fit_transform(X_train)
+    X_val = scaler.transform(X_val)
+
+    # 训练模型
+    final_weights = train_logreg(X_train, y_train, learning_rate=0.01, iterations=5000, batch_size=64, lambda_reg=0.1)
+
+    # 验证模型
+    val_accuracy = accuracy(X_val, y_val, final_weights)
+    print(f"验证集的准确率: {val_accuracy:.4f}")
 
     # 加载测试集
-    X_test, y_test = load_data("./horse-colic.test")
+    X_test, y_test = load_data("./data/horse-colic.test")
+    X_test = scaler.transform(X_test)
 
-    # 训练对率回归模型
-    weights = train_logreg(X_train, y_train)
-
-    # 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]}")
+    # 测试模型
+    final_test_accuracy = accuracy(X_test, y_test, final_weights)
+    print(f"最终测试集的准确率: {final_test_accuracy:.4f}")