更新 function.py

function.py

@@ -1,41 +1,15 @@
-# function.py
 import numpy as np
 import os
 from sklearn.linear_model import LogisticRegression
 from sklearn.preprocessing import StandardScaler
 from sklearn.model_selection import train_test_split
-
-"""
-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。
-
-注意事项:
-- 在 function.py 中实现对率回归算法。
-- 可以自行导入其他库，需要在 requirements.txt 添加依赖，若未在 requirements.txt 中添加使用到的依赖，则无法自动评测，成绩为零。
-- 自动评测脚本仅会加载 train_logreg、predict_logreg 和 load_data 函数，请勿修改函数名。
-
-测评环境:
-- CPU: Intel Xeon(R) CPU E3-1231 v3 @ 3.40GHz
-- OS: Ubuntu 22.04.4 LTS
-- Python: 3.10.12
-- numpy: 2.1.1
-"""
-
-
-
-# 学号和姓名
-id = "202210423204"
-name = "陈垣豪"
-
-import numpy as np
-import os
-from sklearn.preprocessing import StandardScaler
-from sklearn.model_selection import train_test_split
+import joblib  # 用于保存模型
 
 def sigmoid(z):
     """计算sigmoid函数"""
     return 1.0 / (1 + np.exp(-z))
 
+
 def compute_gradient(X_batch, y_batch, weights, lambda_reg):
     """计算梯度，包含L2正则化"""
     z = np.dot(X_batch, weights)
@@ -44,12 +18,13 @@ def compute_gradient(X_batch, y_batch, weights, lambda_reg):
     gradient += (lambda_reg / len(y_batch)) * weights  # L2正则化
     return gradient
 
+
 def train_logreg(X, y, learning_rate=0.001, iterations=10000, batch_size=64, lambda_reg=0.1):
     """逻辑回归训练函数。"""
     m, n = X.shape
     weights = np.random.randn(n) * np.sqrt(2. / n)  # He初始化
     print("开始训练逻辑回归模型")
-    
+
     for i in range(iterations):
         indices = np.random.permutation(m)
         X_shuffled = X[indices]
@@ -60,23 +35,25 @@ def train_logreg(X, y, learning_rate=0.001, iterations=10000, batch_size=64, lam
             y_batch = y_shuffled[j:j + batch_size]
             gradient = compute_gradient(X_batch, y_batch, weights, lambda_reg)
             weights -= learning_rate * gradient
-            
+
         if i % 1000 == 0:  # 每1000次迭代打印一次信息
             print(f"迭代 {i}: 当前权重形状: {weights.shape}")
 
     return weights
 
+
 def predict_logreg(inX, weights):
     """逻辑回归预测函数。"""
     z = np.dot(inX, weights)
     h = sigmoid(z)
     return 1 if h >= 0.5 else 0
 
+
 def load_data(filepath):
     """从文件中加载数据。"""
     if not os.path.exists(filepath):
         raise FileNotFoundError(f"文件 {filepath} 不存在！")
-    
+
     data, labels = [], []
     print(f"加载数据中: {filepath}")
     with open(filepath, 'r') as f:
@@ -89,30 +66,31 @@ def load_data(filepath):
 
     data = np.array(data)
     labels = np.array(labels)
-    
+
     num_features = data.shape[1]
-    indices_to_remove = [2]  
-    
+    indices_to_remove = [2]
+
     if num_features > 23:
         indices_to_remove.extend(range(23, min(28, num_features)))
 
     data = np.delete(data, indices_to_remove, axis=1)
     target_variable = np.where((labels == 2) | (labels == 3), 0, 1)
 
-    print(f"数据加载成功，数据形状: {data.shape}, 标签形状: {target_variable.shape}")
     return data, target_variable
 
+
 def bagging_predict(X, models):
     """集成学习预测函数。"""
     predictions = np.zeros((X.shape[0], len(models)))
-    
+
     for i, model in enumerate(models):
         predictions[:, i] = [predict_logreg(x, model) for x in X]
-    
+
     # 投票
     final_predictions = np.round(np.mean(predictions, axis=1))
     return final_predictions
 
+
 def k_fold_cross_validation(X, y, k=5, learning_rate=0.001, iterations=10000, batch_size=64, lambda_reg=0.1):
     """K折交叉验证。"""
     m = X.shape[0]
@@ -122,7 +100,7 @@ def k_fold_cross_validation(X, y, k=5, learning_rate=0.001, iterations=10000, ba
     for i in range(k):
         val_start = i * fold_size
         val_end = (i + 1) * fold_size if i != k - 1 else m
-        
+
         X_val = X[val_start:val_end]
         y_val = y[val_start:val_end]
         X_train = np.concatenate((X[:val_start], X[val_end:]))
@@ -143,11 +121,11 @@ def k_fold_cross_validation(X, y, k=5, learning_rate=0.001, iterations=10000, ba
 
     return np.mean(accuracies)
 
+
 if __name__ == "__main__":
     # 加载训练集
     X_train, y_train = load_data("./data/horse-colic.data")
 
-
     # 数据归一化处理
     scaler = StandardScaler()
     X_train = scaler.fit_transform(X_train)  # 使用训练集进行标准化
@@ -159,7 +137,6 @@ if __name__ == "__main__":
     # 加载测试集
     X_test, y_test = load_data("./data/horse-colic.test")
 
-
     # 测试集也需要进行相同的归一化
     X_test = scaler.transform(X_test)  # 使用相同的Scaler进行测试集的标准化
 
@@ -176,3 +153,6 @@ if __name__ == "__main__":
     accuracy = np.mean(y_pred == y_test)
     print(f"测试集准确率: {accuracy:.4f}")
 
+    # 保存训练的模型
+    joblib.dump(models, "bagging_models.pkl")
+    print("模型已保存为 bagging_models.pkl")