34Build an ML training set in SQL with point-in-time-correct feature joins (no future leakage).▼hardUberDoorDashDatabricks2 replies◆ premiumThe most frequent way SQL leaks the future into a training set is a sloppy join to a feature table. The fix is point-in-time correctness, implemented as an as-of join. Here is how to write it.Open full answer →
02What is a feature store, and how does it prevent training-serving skew?▼mediumDatabricksMicrosoftAmazon2 repliesunlockedA feature store is simple to define and simple to botch. The signal is naming the precise bug it prevents (training-serving skew) and the point-in-time correctness that blocks label leakage. Here is the full picture.Open full answer →
