上传文件至 /

function.py

@@ -3,7 +3,7 @@ from sklearn.ensemble import RandomForestClassifier
 from sklearn.preprocessing import PolynomialFeatures, StandardScaler
 from sklearn.impute import SimpleImputer
 from sklearn.metrics import accuracy_score
-from sklearn.model_selection import train_test_split
+from sklearn.model_selection import train_test_split, GridSearchCV
 
 id = "202210423104"
 name = "张钦诚"
@@ -38,15 +38,20 @@ def preprocess_data(X):
 
 def train_logreg(X, y):
     X_processed = preprocess_data(X)  # 预处理训练数据
-    model = RandomForestClassifier(
-        n_estimators=300,  # 增加树的数量
-        max_depth=25,      # 调整最大深度
-        min_samples_split=5,
-        class_weight='balanced',  # 处理类别不平衡
-        random_state=42
-    )
-    model.fit(X_processed, y)
-    return model  # 返回训练好的模型
+
+    # 超参数调优
+    param_grid = {
+        'n_estimators': [100, 200, 300],
+        'max_depth': [None, 15, 25],
+        'min_samples_split': [2, 5, 10],
+        'class_weight': ['balanced', None]
+    }
+
+    model = RandomForestClassifier(random_state=42)
+    grid_search = GridSearchCV(model, param_grid, cv=5, scoring='accuracy', n_jobs=-1)
+    grid_search.fit(X_processed, y)
+
+    return grid_search.best_estimator_  # 返回最佳模型
 
 
 def predict_logreg(X, model):
@@ -57,7 +62,6 @@ def predict_logreg(X, model):
 
 
 if __name__ == "__main__":
-    # 数据加载和预处理保持不变
     X, y = load_data("./data/horse-colic.data")
 
     # 划分训练集和测试集
@@ -71,4 +75,4 @@ if __name__ == "__main__":
 
     # 计算准确率
     accuracy = accuracy_score(y_test, y_pred)
-    print(f"Accuracy: {accuracy:.2f}")
+    print(f"Accuracy: {accuracy:.2f}")