On the Need for a Language Describing Distribution Shifts: Illustrations on Tabular Datasets

Open in new window