机器学习分类使用交叉验证时怎么设置分层抽样？

进行机器学习分类使用交叉验证，无论选择哪种算法，都会分析失败， Traceback (most recent call last): File "internal\service\analysis_service\helper.py", line 550, in invoke File "internal\web\handler\analysis_handler\supervised_learning\classifier\gbdt_classifier.py", line 264, in do_analysis File "seal_algorithm\render\render\supervised_learning\classifier\gbdt_classifier.py", line 144, in initFile "seal_algorithm\render\render\supervised_learning\classifier\gbdt_classifier.py", line 20, in initFile "seal_algorithm\algorithm\supervised_learning\classifier\gbdt_classifier.py", line 104, in initFile "seal_algorithm\algorithm\supervised_learning\classifier\gbdt_classifier.py", line 240, in cross_validation_main File "sklearn\utils\validation.py", line 63, in inner_f File "sklearn\metrics_ranking.py", line 547, in roc_auc_score File "sklearn\metrics_base.py", line 119, in _average_binary_score File "sklearn\metrics_ranking.py", line 327, in _binary_roc_auc_score ValueError: Only one class present in y_true. ROC AUC score is not defined in that case.

猜测没有分层抽样，在哪里设置分层抽样？？？？？？