feat: persist leftover observed Y and expected E (v2.12.20) - #526
feat: persist leftover observed Y and expected E (v2.12.20)#526seonghobae wants to merge 2 commits into
Conversation
Name leftover_observed_score and leftover_expected_score on leftover pairs so leftover residual R = Y − E is checkable (ADR 0168). Missing cells stay out of the factorization. Do not invent a leftover score or a second theta.
|
Warning Review limit reachedNext included review available in 2 minutes. View limit detailsLimit details: You’ve used the included review currently available. You've used all free OSS reviews for now. Wait for the free limit to reset to keep reviewing this public repository. Review configuration: ⚙️ Run configurationConfiguration used: Organization UI Review profile: CHILL Plan: Pro Plus Run ID: ⛔ Files ignored due to path filters (1)
📒 Files selected for processing (21)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![devin-ai-integration[bot]](https://avatars.githubusercontent.com/in/811515?s=60&v=4){kind=small}
| return ( | ||
| LeftoverPair( | ||
| pair_kind=PAIR_KIND_CLOSEST, | ||
| post_id=closest[1], | ||
| criterion_code=closest[2], | ||
| leftover_distance=closest[0], | ||
| leftover_residual=closest[3], | ||
| ), | ||
| LeftoverPair( | ||
| pair_kind=PAIR_KIND_FARTHEST, | ||
| post_id=farthest[1], | ||
| criterion_code=farthest[2], | ||
| leftover_distance=farthest[0], | ||
| leftover_residual=farthest[3], | ||
| ), | ||
| max(distance, 0.0), | ||
| post_ids[person], | ||
| item_codes[item], | ||
| float(residual[person, item]), | ||
| float(matrix[person, item]), | ||
| float(expected[person, item]), | ||
| ) |
There was a problem hiding this comment.
📝 Info: Residual identity check is float-safe
The check constraint leftover_pair_residual_identity requires abs(R-(Y-E))<1e-9. In leftover_pairs.py, R is matrix-expected while Y and E are the same operands, so R and Y-E differ only by float epsilon (~1e-16). Rebuild/seed inserts cannot spuriously fail the constraint.
Was this helpful? React with 👍 or 👎 to provide feedback.
|
Do not merge this PR onto protected This head (
The independent reconstruction onto protected |
…pair login build migrations/0001_initial_schema.sql is shipped, baked into the Postgres image for fresh installs only. This PR had duplicated the new leftover_observed_score/leftover_expected_score columns (plus their residual-identity check constraint) directly into 0001 as well as into the correctly idempotent 0108_report_leftover_observed_expected.sql migration. Reverts 0001 to its shipped form (byte-identical to origin/main) so 0108 is the single source on every install path, and wires 0108 into the two hand-enumerated migration fixtures (tests/test_schema.py::schema_db, backend/tests/test_api.py::seeded_db) so removing it from 0001 doesn't silently stop creating the columns. Also repairs this branch's build: the login button still built an unsanitized returnUrl inline instead of using returnUrlFromLocation()/rememberOidcReturnUrl(), and an unreachable login-screen AdminPanel render (accessToken is string | undefined pre-login) failed the TypeScript build. Same root cause and fix as every other branch that inherited this from before the canonical fix landed on main.
1 participant
Buyer next action
After
make seed, leftover closest/farthest pairs still sit above the period-report member list. Each pair now names observed scoreY, expected scoreE[Y|θ, item], leftover residualR = Y − E, and leftover-map distanced. Read those badges, check leftover residual equals observed minus expected, then click the pair to open that post.Never invent a leftover score. Never invent a theta.
Exact unique scope
Independent leftover honesty slice on protected
main@ef6f5a5(v2.12.6). Exact head:60ec4e1e0731f624c5c1a75710fbdf82776c3f94.leftover_observed_scoreandleftover_expected_scoreonreport_leftover_pair(3NF, two-or-more-wordsnake_case).leftover_pair_residual_identity:abs(R − (Y − E)) < 1e-9.0108upgrades existing volumes. Existing leftover rows are derived; incomplete rows are deleted so rebuild/seed rewrites honestYandEinstead of fabricating them.GET /api/reports/{grouping}/{period}returnsYandE.Why this is independent
Does not mix into leftover persist-map (#481), leftover criterion landing (#485), leftover residual UI (#502), leftover complete-case coverage (#518), leftover-map axis share (#519), leftover comparison-strip (#521), or two-axis leftover-map distance (#522). Closed #511 shipped Y/E onto the unprotected
#426stack, not protectedmain.Issues #79 and #87 stay open.
Merge contract
Independent exact-head APPROVE required. Do not self-approve. Squash-merge only with independent APPROVE + exact-head required success.
Tests
uv run pytest tests/test_leftover_pairs.py tests/test_migration_replay.py: 8 passed (Y=4/1, E=3/2 residual identity; observed ≠ residual; 0108 replay).pnpm exec vitest run src/App.test.tsx: 78 passed, including observed/expected/leftover badges and next-action copy.Papers
Gabriel, K. R. (1971). The biplot graphic display of matrices with application to principal component analysis. Biometrika, 58(3), 453–467. https://doi.org/10.1093/biomet/58.3.453
Jeon, M., Jin, I. H., Schweinberger, M., & Baugh, S. (2021). Mapping unobserved item–respondent interactions: A latent space item response model with interaction map. Psychometrika, 86(2), 378–403. https://doi.org/10.1007/s11336-021-09762-5