You work for a retail company. You have a managed tabular dataset in Vertex AI that contains sales data from three different stores. The dataset includes several features, such as store name and sale

Question

Options

• AUse Vertex AI manual split, using the store name feature to assign one store for each set
• BUse Vertex AI default data split
• CUse Vertex AI chronological split, and specify the sales timestamp feature as the time variable
• DUse Vertex AI random split, assigning 70% of the rows to the training set, 10% to the validation

• A
  4% (1)
• B
  8% (2)
• C
  76% (19)
• D
  12% (3)

Explanation

Sales data is temporal: future sales depend on past trends, seasonality, and patterns that unfold over time. A chronological (time-based) split - training on older data, validating and testing on progressively newer data - prevents temporal data leakage (where the model sees future information during training) and produces a realistic estimate of how the model performs on unseen future data, including at the new store. Option A (split by store name, one store per set) would mean the model has never seen any data from the test store, making the test set unrepresentative. Options B and D (default or random splits) ignore the temporal structure and allow future data points to appear in the training set, inflating performance estimates.

No community discussion yet for this question.