“混合模型”(例如，包括KerasWrapper的SKlearn管道)的模型日志记录可能吗？

from keras.wrappers.scikit_learn import KerasClassifier
from sklearn.pipeline import Pipeline
from sklearn.model_selection import cross_val_score
from sklearn.preprocessing import StandardScaler
import logging

def create_model():
    # 在此定义Keras模型结构
    model = ...
    return model

# 创建KerasWrapper的SKlearn管道
estimators = []
estimators.append(('standardize', StandardScaler()))
estimators.append(('mlp', KerasClassifier(build_fn=create_model, epochs=10, batch_size=32, verbose=0)))

pipeline = Pipeline(estimators)

# 设置日志记录
logging.basicConfig(filename='model.log', level=logging.INFO)

# 训练和评估模型
scores = cross_val_score(pipeline, X, y, cv=5)
logging.info('模型评估结果：{}'.format(scores))