333

function.py

@@ -1,7 +1,5 @@
 # function.py
 import numpy as np
-from sklearn.preprocessing import StandardScaler
-
 """
 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。
@@ -22,56 +20,148 @@ from sklearn.preprocessing import StandardScaler
 """
 请在这里填写学号和姓名，不填写即成绩无效。
 """
-id = "202210423221"
-name = "廖子豪"
+id = "=202210423221"
+name = "=廖子豪"
 
 
 # 这里写自己的代码实现
-def load_data(filepath):
-    try:
-        data = np.genfromtxt(filepath, delimiter=' ', dtype=float, filling_values=np.nan)
-        print(f"Loaded data shape: {data.shape}")
-        if data.ndim == 1:
-            raise ValueError("数据加载后为一维数组，预期为二维数组。请检查数据文件格式。")
-        if data.shape[1] < 2:
-            raise ValueError("数据文件中没有足够的列。每行至少需要包含一个特征和一个标签。")
-        X = data[:, :-1]
-        y = data[:, -1]
-        X = np.nan_to_num(X, nan=np.nanmean(X, axis=0))
-        return X, y
-    except Exception as e:
-        print(f"Error loading data: {e}")
-        return np.empty((0, 0)), np.empty(0)
+def train_logreg(X, y, numIter=1000, learning_rate=0.06, batch_size=128, lambda_reg=0.01):
+    """
+    对率回归训练函数，加入L2正则化。
+
+    Args:
+    - X (numpy array): 特征数据。
+    - y (numpy array): 标签数据。
+    - numIter (int): 迭代次数。默认为1000。
+    - learning_rate (float): 学习率。默认为0.06。
+    - batch_size (int): 批量大小。默认为128。
+    - lambda_reg (float): L2正则化参数。默认为0.01。
+
+    Returns:
+    - numpy array: 训练得到的权重。
+    """
+def train_logreg(X, y, numIter=1000, learning_rate=0.06, batch_size=128, lambda_reg=0.01):
+    """
+    使用梯度上升的对率回归训练函数，加入L2正则化。
+
+    Args:
+    - X (numpy array): 特征数据。
+    - y (numpy array): 标签数据。
+    - numIter (int): 迭代次数。默认为1000。
+    - learning_rate (float): 学习率。默认为0.06。
+    - batch_size (int): 批量大小。默认为128。
+    - lambda_reg (float): L2正则化参数。默认为0.01。
+
+    Returns:
+    - numpy array: 训练得到的权重。
+    """
+    m, n = X.shape  # m: 样本数, n: 特征数
+    weights = np.zeros(n)  # 参数初始化
+
+    for j in range(numIter):
+        # 随机打乱样本顺序
+        dataIndex = np.random.permutation(m)
+        for i in range(0, m, batch_size):
+            end = min(i + batch_size, m)
+            # 批量梯度上升
+            batch_X = X[dataIndex[i:end]]
+            batch_y = y[dataIndex[i:end]]
+            
+            # 计算梯度
+            z = np.dot(batch_X, weights)
+            h = 1 / (1 + np.exp(-z))  # 计算h
+            error = batch_y - h  # 计算误差
+            gradient = np.dot(batch_X.T, error) / len(batch_y)  # 计算梯度
+            # 添加L2正则化
+            reg_term = 2 * lambda_reg * weights
+            gradient -= reg_term / len(batch_y)
+            weights += learning_rate * gradient  # 更新权重
 
-def train_logreg(X, y, learning_rate=0.01, iterations=1000):
-    m, n = X.shape
-    X = np.c_[np.ones((m, 1)), X]  # 添加偏置项
-    weights = np.zeros(n + 1)
-    for _ in range(iterations):
-        for i in range(m):
-            z = np.dot(X[i], weights)
-            y_pred = 1 / (1 + np.exp(-z))
-            weights -= learning_rate * (y[i] - y_pred) * X[i]
     return weights
 
+
+
+
 def predict_logreg(inX, weights):
-    if inX.ndim == 1:
-        inX = inX.reshape(1, -1)
-    inX = np.c_[np.ones((1, 1)), inX]  # 添加偏置项
-    z = np.dot(inX, weights)
-    y_pred = 1 / (1 + np.exp(-z))
-    return 1 if y_pred > 0.5 else 0
+    """
+    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)  # 线性组合
+    prob = 1 / (1 + np.exp(-z))  # 计算概率
+    return 1 if prob >= 0.5 else 0  # 返回分类标签
+
+
+
+# 这里也写自己的代码实现
+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**。
+
+    Args:
+    - directory (str): 数据文件路径。
+
+    Returns:
+    - np.array: 特征数据。
+    - np.array: 标签数据。
+    """
+
+    file = open(filepath)
+    fr = file.readlines()
+    data_list = []
+    y = []
+    for data in fr:
+        data = data.strip()
+        data = data.split(' ')
+        data_25 = data[24]
+        y.append(data[22])
+        del data[-1]
+        del data[26]
+        del data[25]
+        del data[24]
+        del data[22]
+        del data[2]
+        for i in data_25:
+            data.append(i)
+        data = [0 if data_elem == '?' else float(data_elem) for data_elem in data]
+        data_list.append(data)
+    data_number = len(y)
+    y_train = np.array([1 if y[j] == '2' or y[j] == '3' else 0 for j in range(data_number)])
+    X = np.array(data_list)
+    X_train = np.insert(X, 0, 1, axis=1)
+    return X_train, y_train
+   
+
+
 
 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("./data/horse-colic.data")
+
+    # 加载测试集
     X_test, y_test = load_data("./data/horse-colic.test")
 
-    scaler = StandardScaler()
-    X_train = scaler.fit_transform(X_train)
-    X_test = scaler.transform(X_test)
-
+    # 训练对率回归模型
     weights = train_logreg(X_train, y_train, learning_rate=0.01, iterations=1000)
 
-    y_pred_all = [predict_logreg(X_test[i], weights) for i in range(len(X_test))]
-    accuracy = np.mean(y_pred_all == y_test)
-    print(f"Accuracy: {accuracy}")
+    # 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]}")'''
+    X_train, y_train = load_data("./data/horse-colic.data")
+    X_test, y_test = load_data("./data/horse-colic.test")
+    weights = train_logreg(X_train, y_train, numIter=1000, learning_rate=0.01,lambda_reg=0.2)
+    error=0
+    for i in range(len(y_test)):
+        y_pred = predict_logreg(X_test[i], weights)
+        if y_pred != y_test[i]:
+            error+=1
+        print(f"预测结果: {y_pred}, 实际结果: {y_test[i]}")
+    print((len(y_test)-error)/len(y_test))