Title: feat: Opt-in `created_timestamp <= entity_timestamp` cutoff in point-in-time joins (as-of-known-time retrieval) · Issue #6615 · feast-dev/feast · GitHub
Open Graph Title: feat: Opt-in `created_timestamp <= entity_timestamp` cutoff in point-in-time joins (as-of-known-time retrieval) · Issue #6615 · feast-dev/feast
X Title: feat: Opt-in `created_timestamp <= entity_timestamp` cutoff in point-in-time joins (as-of-known-time retrieval) · Issue #6615 · feast-dev/feast
Description: Is your feature request related to a problem? Please describe. Picking this back up from a Slack thread a while ago, where the dedup-only handling of created_timestamp_column was confirmed. get_historical_features enforces point-in-time ...
Open Graph Description: Is your feature request related to a problem? Please describe. Picking this back up from a Slack thread a while ago, where the dedup-only handling of created_timestamp_column was confirmed. get_his...
X Description: Is your feature request related to a problem? Please describe. Picking this back up from a Slack thread a while ago, where the dedup-only handling of created_timestamp_column was confirmed. get_his...
Opengraph URL: https://github.com/feast-dev/feast/issues/6615
X: @github
Domain: github.com
{"@context":"https://schema.org","@type":"DiscussionForumPosting","headline":"feat: Opt-in `created_timestamp \u003c= entity_timestamp` cutoff in point-in-time joins (as-of-known-time retrieval)","articleBody":"**Is your feature request related to a problem? Please describe.**\n\nPicking this back up from [a Slack thread a while ago](https://feastopensource.slack.com/archives/C01MSKCMB37/p1744885223539189), where the dedup-only handling of `created_timestamp_column` was confirmed.\n\n`get_historical_features` enforces point-in-time correctness on the event timestamp only. When a feature view has a `created_timestamp_column`, it is used purely as a dedup tiebreaker (`MAX(created_timestamp)` per event timestamp), never as a filter. There is no `created_timestamp \u003c= entity_timestamp` predicate anywhere in the join templates (verified on current master, BigQuery template as reference; the other offline stores follow the same pattern).\n\nAs a result, a training row at entity timestamp `T` can be served a value that did not exist yet at `T`: its `event_timestamp \u003c= T` but its `created_timestamp > T` (backfills, late corrections, delayed batch writes). This causes:\n\n- Feature leakage: training rows see values that were written after the event.\n- Non-reproducible training sets: re-running after a backfill silently changes past feature vectors.\n- Offline/online skew: the online model at `T` served the older value, offline retrieval returns the newer one.\n\nConcrete example: two rows for the same entity and event timestamp, value `4000` created before the entity timestamp and `6000` created after it. `get_historical_features` returns `6000`.\n\n**Describe the solution you'd like**\n\nAn opt-in, per-query flag, default off:\n\n```python\nstore.get_historical_features(\n entity_df=...,\n features=...,\n filter_by_created_timestamp=True, # only use feature versions known as of each entity row's timestamp\n)\n```\n\nWhen enabled and the feature view has a `created_timestamp_column`, add a single predicate to the point-in-time join:\n\n```sql\nAND subquery.created_timestamp \u003c= entity_dataframe.entity_timestamp\n```\n\nDefault `False` produces identical SQL and results to today. The flag can be threaded through the same optional-kwargs mechanism as the recently added `start_date`/`end_date`, so offline stores that have not implemented it are unaffected unless a user explicitly opts in.\n\n**Describe alternatives you've considered**\n\n- A per-`FeatureView` or config-level setting. Per-query is preferred since the same view may want both leak-free training and \"latest known\" analysis.\n- #2980 proposed a global `max_created_time` fixed at query time (closed by the stale bot). This proposal is the per-row variant: each entity row is cut off at its own timestamp, which is what you need to emulate what the online store would have served at each event.\n- We currently run a patched BigQuery join template with this predicate in production. It works but has to be re-synced on every Feast upgrade.\n\n**Additional context**\n\n- Rows with a NULL `created_timestamp` would be excluded by the predicate; suggest documenting the column must be non-null when the mode is enabled.\n- Orthogonal to `ttl`: TTL bounds by event time, this bounds by known time.\n- I have a PR ready implementing this across the offline stores (SQL templates, ibis-based stores, dask) with tests. Happy to iterate on naming and API surface.\n","author":{"url":"https://github.com/addenergyx","@type":"Person","name":"addenergyx"},"datePublished":"2026-07-19T16:32:07Z","url":"https://github.com/feast-dev/feast/issues/6615"}
| react-profiling | 0 |
| react-import-map | react,react-dom,react-dom/client,react-dom/profiling,react-is,react/compiler-runtime,react/jsx-dev-runtime,react/jsx-runtime,scheduler@b8488ee2c322 |
| route-pattern | /_view_fragments/issues/show/:user_id/:repository/:id/issue_layout(.:format) |
| route-controller | voltron_issues_fragments |
| route-action | issue_layout |
| fetch-nonce | v2:ad2c5f1d-201d-334e-5d11-f677c82fa13a |
| current-catalog-service-hash | 81bb79d38c15960b92d99bca9288a9108c7a47b18f2423d0f6438c5b7bcd2114 |
| request-id | AFE6:EFEB1:F4FAE1:14236C9:6AFE981F |
| html-safe-nonce | a6cfb2a3538569ec093315ad89a2701db506e7ec41c4c5952a2a843a551a5830 |
| visitor-payload | eyJyZWZlcnJlciI6IiIsInJlcXVlc3RfaWQiOiJBRkU2OkVGRUIxOkY0RkFFMToxNDIzNkM5OjZBRkU5ODFGIiwidmlzaXRvcl9pZCI6IjIwOTk0OTA5MjA3MjUzODcyOTUiLCJyZWdpb25fZWRnZSI6ImlhZCIsInJlZ2lvbl9yZW5kZXIiOiJpYWQifQ== |
| visitor-hmac | b1b97f051f2e21b692d10cd60b6187b5c8f5a81fe749783da03f3019c03118f6 |
| hovercard-subject-tag | issue:4923485884 |
| github-keyboard-shortcuts | repository,issues,copilot |
| google-site-verification | Apib7-x98H0j5cPqHWwSMm6dNU4GmODRoqxLiDzdx9I |
| octolytics-url | https://collector.github.com/github/collect |
| analytics-location | / |
| fb:app_id | 1401488693436528 |
| apple-itunes-app | app-id=1477376905, app-argument=https://github.com/_view_fragments/issues/show/feast-dev/feast/6615/issue_layout |
| twitter:image | https://opengraph.githubassets.com/0fa179d4e15922ac5c9b2c39d806ea067c8d0c5eb3a14d17b566db292f4bae44/feast-dev/feast/issues/6615 |
| twitter:card | summary_large_image |
| og:image | https://opengraph.githubassets.com/0fa179d4e15922ac5c9b2c39d806ea067c8d0c5eb3a14d17b566db292f4bae44/feast-dev/feast/issues/6615 |
| og:image:alt | Is your feature request related to a problem? Please describe. Picking this back up from a Slack thread a while ago, where the dedup-only handling of created_timestamp_column was confirmed. get_his... |
| og:image:width | 1200 |
| og:image:height | 600 |
| og:site_name | GitHub |
| og:type | object |
| og:author:username | addenergyx |
| hostname | github.com |
| expected-hostname | github.com |
| None | b3c0f155eafc27e60db6e254d1ffff0e072e9ab3b4018a5ee518797c19d89f86 |
| turbo-cache-control | no-preview |
| go-import | github.com/feast-dev/feast git https://github.com/feast-dev/feast.git |
| octolytics-dimension-user_id | 57027613 |
| octolytics-dimension-user_login | feast-dev |
| octolytics-dimension-repository_id | 161133770 |
| octolytics-dimension-repository_nwo | feast-dev/feast |
| octolytics-dimension-repository_public | true |
| octolytics-dimension-repository_is_fork | false |
| octolytics-dimension-repository_network_root_id | 161133770 |
| octolytics-dimension-repository_network_root_nwo | feast-dev/feast |
| turbo-body-classes | logged-out env-production page-responsive |
| disable-turbo | false |
| browser-stats-url | https://api.github.com/_private/browser/stats |
| browser-errors-url | https://api.github.com/_private/browser/errors |
| release | 4a10310af51a1bfffbedc4fd81cb582acf0cb0ff |
| ui-target | full |
| theme-color | #1e2327 |
| color-scheme | light dark |
| react-app-name | issues |
Links:
Viewport: width=device-width