数据泄露风险
Author: OstrichThinkingCreated Jan 8, 2026Updated Sep 8, 2026
在进行分类模型训练时:
def train(self, setting): train_data, train_loader = self._get_data(flag='TRAIN') vali_data, vali_loader = self._get_data(flag='TEST') test_data, test_loader = self._get_data(flag='TEST')
def test(self, setting, test=0): test_data, test_loader = self._get_data(flag='TEST') if test:
为训练集、测试集分别初始化 Dataloader, 每次初始化涉及到标准化操作:
pre_process
normalizer = Normalizer() self.feature_df = normalizer.normalize(self.feature_df) print(len(self.all_IDs))
是否存在“训练集用自己均值方差进行标准化、测试集也用自己的均值方差进行标准化”的问题?
Source: thuml/Time-Series-Library