AppliedAIPrep logoAppliedAI/Prep
← SQL & Data Engineering / 38
mediumMetaAmazonDatabricks

Pull a random sample, and a stratified sample, for an ML training set in SQL.

Building a training set from a billion-row table means sampling, and the naive 'ORDER BY random() LIMIT n' sorts the whole table. Stratified sampling needs more care still. Here are both, done efficiently.

Updated Aug 2026 · Grounded in real Applied AI Engineer interview loops and written to a senior-engineer editorial bar.

Building a training set from a billion-row table means sampling, and the naive 'ORDER BY random() LIMIT n' sorts the whole table. Stratified sampling needs more care still. Here are both, done efficiently.

Unlock the other 750 answers · ₹2,000 / $25Your progress and mastery stay saved · 6 months · one payment · no auto-renew
UP NEXT ON YOUR JOURNEY
Next in this trackGenerate ML labels in SQL: did each user churn (no activity in the next 30 days)?Next in this trackBin a continuous feature into quantile and fixed-width buckets in SQL. When do you use each?Next in this trackDetect feature drift in SQL: compare a feature's distribution between two time periods.
DISCUSSION · 0

No comments yet — be the first to share your approach.