如何在sklearn中调优score_func (feature selection)的超参数？

from sklearn.feature_selection import SelectKBest
from sklearn.model_selection import GridSearchCV
from sklearn.metrics import make_scorer
from sklearn.pipeline import Pipeline
from sklearn.svm import SVC

def custom_score_func(X, y):
    # 自定义评估函数，根据实际需求编写
    # 返回一个评估指标，例如准确率、F1分数等
    pass

feature_selector = SelectKBest(score_func=custom_score_func)
classifier = SVC()
pipeline = Pipeline([('feature_selector', feature_selector), ('classifier', classifier)])

param_grid = {
    'feature_selector__k': [5, 10, 15],  # 特征选择器选择的特征数量
    'classifier__C': [0.1, 1, 10],  # SVM分类器的正则化参数
    'classifier__kernel': ['linear', 'rbf']  # SVM分类器的核函数
}

grid_search = GridSearchCV(pipeline, param_grid=param_grid, scoring=make_scorer(custom_score_func))
grid_search.fit(X, y)

print("Best parameters: ", grid_search.best_params_)
print("Best score: ", grid_search.best_score_)