更新 function.py

function.py

@@ -33,7 +33,6 @@ name = "邓龙树"
 
 
 # 这里写自己的代码实现
-
 # 数据加载函数保持不变
 def load_data(filepath):
     """
@@ -76,15 +75,6 @@ def load_data(filepath):
     return data, labels
 
 
-# 封装训练和评估模型的函数
-def train_and_evaluate_model(X_train_fold, y_train_fold, X_val_fold, y_val_fold, C):
-    model = SVC(kernel='rbf', C=C, gamma='scale', random_state=42)
-    model.fit(X_train_fold, y_train_fold)
-    y_pred_fold = model.predict(X_val_fold)
-    accuracy_fold = accuracy_score(y_val_fold, y_pred_fold)
-    return accuracy_fold, model
-
-
 def train_svm(X, y, other_params=None):
     """
     支持向量机训练函数。
@@ -124,21 +114,30 @@ if __name__ == "__main__":
     # 划分数据集（这里划分出一部分作为最终测试集，剩下的用于交叉验证训练）
     X_full_train, X_final_test, y_full_train, y_final_test = train_test_split(X_train, y_train, test_size=0.2, random_state=42, stratify=y_train)
 
-    # 数据标准化
-    scaler = StandardScaler()
+    # 数据标准化，使用MinMaxScaler进行归一化
+    scaler = MinMaxScaler()
     X_full_train = scaler.fit_transform(X_full_train)
     X_final_test = scaler.transform(X_final_test)
 
-       # 定义要搜索的超参数网格
-    param_grid = {
-        'C': [0.1, 1, 10],
-        'gamma': [0.1, 1, 10]
-    }
-    svc = SVC(kernel='rbf', random_state=42)
-    grid_search = GridSearchCV(svc, param_grid, cv=5, scoring='accuracy')
-    grid_search.fit(X_full_train, y_full_train)
-
-    best_model = grid_search.best_estimator_
+    # 定义要尝试的C值列表
+    C_values = [0.1, 1, 10]
+    best_accuracy = 0
+    best_model = None
+    kf = KFold(n_splits=5, shuffle=True, random_state=42)
+    for C in C_values:
+        total_accuracy = 0
+        for train_index, val_index in kf.split(X_full_train):
+            X_train_fold, X_val_fold = X_full_train[train_index], X_full_train[val_index]
+            y_train_fold, y_val_fold = y_full_train[train_index], y_full_train[val_index]
+            model = SVC(kernel='rbf', C=C, gamma='scale', random_state=42)
+            model.fit(X_train_fold, y_train_fold)
+            y_pred_fold = model.predict(X_val_fold)
+            accuracy_fold = accuracy_score(y_val_fold, y_pred_fold)
+            total_accuracy += accuracy_fold
+        avg_accuracy = total_accuracy / 5
+        if avg_accuracy > best_accuracy:
+            best_accuracy = avg_accuracy
+            best_model = model
 
     # 使用在交叉验证中选出的最佳模型在最终测试集上进行评估
     y_pred = best_model.predict(X_final_test)