如何在scikit-learn管道的各个步骤之间传递值？

from sklearn.pipeline import Pipeline
from sklearn.preprocessing import StandardScaler
from sklearn.decomposition import PCA
from sklearn.linear_model import LogisticRegression

# 自定义转换器，用于在步骤之间传递值
class CustomTransformer:
    def __init__(self, value):
        self.value = value
    
    def fit(self, X, y=None):
        return self
    
    def transform(self, X):
        # 在这里可以使用self.value对数据进行处理
        transformed_X = X + self.value
        return transformed_X

# 创建Pipeline对象
pipeline = Pipeline([
    ('custom_transformer', CustomTransformer(value=10)),
    ('scaler', StandardScaler()),
    ('pca', PCA(n_components=2)),
    ('classifier', LogisticRegression())
])

# 使用Pipeline进行训练和预测
pipeline.fit(X_train, y_train)
y_pred = pipeline.predict(X_test)