更新 function.py

function.py

@@ -32,38 +32,53 @@ from sklearn.model_selection import train_test_split
 from sklearn.linear_model import LogisticRegression  
 from sklearn.metrics import accuracy_score, classification_report  
 
-# 读取数据  
-data_path = './data/horse-colic.data'  
-data = pd.read_csv(data_path, header=None, delim_whitespace=True, na_values='?')  
+def load_data(file_path):
+    # 读取数据
+    data = pd.read_csv(file_path, header=None, delim_whitespace=True, na_values='?')
+    # 处理缺失值（用0填充）
+    data.fillna(0, inplace=True)
+    
+    # 选择特征和目标变量
+    # 去掉特征3和特征24~28
+    features = data.iloc[:, :-5].values  # 假设最后一列是目标变量，前面的都是特征
+    target = data.iloc[:, -1].values
 
-# 处理缺失值（用0填充）  
-data.fillna(0, inplace=True)  
+    # 将目标变量中的2和3合并为0
+    target[target == 2] = 0  
+    target[target == 3] = 0  
 
-# 选择特征和目标变量  
-# 去掉特征3和特征24~28  
-features = data.iloc[:, :-5].values  # 假设最后一列是目标变量，前面的都是特征  
-target = data.iloc[:, -1].values  
+    # 将目标变量转换为二进制分类（0或1）
+    target = (target == 1).astype(int)
 
-# 将目标变量中的2和3合并为0  
-target[target == 2] = 0  
-target[target == 3] = 0  
+    return features, target
 
-# 将目标变量转换为二进制分类（0或1）  
-target = (target == 1).astype(int)  
+def train_logreg(X_train, y_train):
+    # 训练逻辑回归模型
+    model = LogisticRegression(max_iter=1000)
+    model.fit(X_train, y_train)
+    return model
 
-# 划分训练集和测试集  
-X_train, X_test, y_train, y_test = train_test_split(features, target, test_size=0.2, random_state=42)  
+def predict_logreg(model, X_test):
+    # 进行预测
+    return model.predict(X_test)
 
-# 训练对率回归模型  
-model = LogisticRegression(max_iter=1000)  
-model.fit(X_train, y_train)  
+# 主程序
+data_path = './data/horse-colic.data'
+features, target = load_data(data_path)
 
-# 预测并评估模型  
-y_pred = model.predict(X_test)  
-accuracy = accuracy_score(y_test, y_pred)  
-report = classification_report(y_test, y_pred)  
+# 划分训练集和测试集
+X_train, X_test, y_train, y_test = train_test_split(features, target, test_size=0.2, random_state=42)
+
+# 训练模型
+model = train_logreg(X_train, y_train)
+
+# 预测并评估模型
+y_pred = predict_logreg(model, X_test)
+accuracy = accuracy_score(y_test, y_pred)
+report = classification_report(y_test, y_pred)
 
 print(f"Accuracy: {accuracy:.2f}")  
 print("Classification Report:")  
 print(report)  
 
+