learning-it by example

Back to the video

Feature Stores: When a Table and a Join Are Enough

On made-up data, a naive join scored a churn rule at 0.96; point in time, the same rule scored 0.69.

Video #159 ยท MLOps and deployment

Steps in the video

  1. 0:00 A naive join scored 0.96. Point in time, the same rule scored 0.69.
  2. 0:39 The idea
  3. 1:07 Setup
  4. 1:28 The made-up data
  5. 1:58 The latest table
  6. 2:13 The naive join
  7. 2:25 The point-in-time join
  8. 2:44 A one-line model
  9. 3:40 The online lookup
  10. 4:09 When a table and a join are enough
  11. 4:43 What this does not show
  12. 5:12 Mistakes to avoid
  13. 5:35 My take and next steps
  14. 5:58 Recap

Files

  • data.py (24 lines)
  • joins.py (14 lines)
  • rule.py (10 lines)
  • run.py (32 lines)

Download the code (zip)

Tested with

Python 3.12.15

What this does not show

Made-up users with a leak I built in on purpose, one seed, a one-line rule scored on the rows it was picked on; no feature store is run, nothing is timed, and vendor claims come from their own docs.

Sources

Learning IT by Example.