2026-09-08 - 2026-09-11

Overview

0 Active Pull Requests
34 Active Issues
Excluding merges, 1 author has pushed 2 commits to main and 45 commits to all branches. On main, 2 files have changed and there have been 102 additions and 12 deletions.

2 Releases published by 1 user

Published v0.9.0 2026-09-10 05:50:45 +00:00

Published v0.8.0 2026-09-08 19:18:56 +00:00

33 Issues closed from 1 user

Closed #74 InferenceError: stringly-typed discriminators, and #[non_exhaustive] on the enum buys nothing while the variants lack it 2026-09-10 05:31:35 +00:00

Closed #53 Declarative competitor configuration: HistoryBuilder::default_rating_for 2026-09-10 05:17:18 +00:00

Closed #52 The joint precision matrix is 0.19% dense and we factorise it densely 2026-09-10 05:08:22 +00:00

Closed #72 Ergonomics: the first program a new user writes does not compile, and the fix is not discoverable 2026-09-10 04:49:47 +00:00

Closed #78 predict_* returns Ok on a NaN-poisoned fit, and eight # Errors sections do not match the code 2026-09-10 04:40:36 +00:00

Closed #75 Vocabulary: "rating" means three things, "sigma" means three things, and "agent" is a fourth word for competitor 2026-09-09 21:23:17 +00:00

Closed #73 Six exported types are unreachable from History — removing them shrinks the API for free 2026-09-09 21:18:59 +00:00

Closed #71 Gaussian's operators are EP internals wearing arithmetic's clothes, and / can return a confident wrong number 2026-09-09 21:13:14 +00:00

Closed #69 Game has no public constructor — every Game::* returns OwnedGame 2026-09-09 20:55:10 +00:00

Closed #77 Docs: no README quickstart, 112 missing_docs warnings, and the only Joint example does not compile downstream 2026-09-09 19:53:56 +00:00

Closed #76 History has no Debug, and trait impls are inconsistent across the public value types 2026-09-09 19:29:20 +00:00

Closed #67 A dropped EventBuilder silently ingests nothing — no #[must_use] 2026-09-09 19:24:40 +00:00

Closed #66 log_evidence_for silently returns whole-history evidence for an unknown key 2026-09-09 19:13:42 +00:00

Closed #70 Accessors disagree on how they report an unknown key, and learning_curve cannot distinguish a typo from a registered competitor 2026-09-09 19:13:42 +00:00

Closed #68 The Time generic is unreachable downstream — Untimed and custom timestamps cannot be used at all 2026-09-09 18:21:58 +00:00

Closed #65 ConstantDrift(-x) is bit-identical to ConstantDrift(x) — seal the field 2026-09-09 17:14:39 +00:00

Closed #62 posterior_of is nondeterministic across processes: float sums over a HashMap 2026-09-09 16:10:52 +00:00

Closed #64 tests/determinism.rs never exercises the parallel sweep it exists to prove 2026-09-09 16:10:52 +00:00

Closed #63 Three production sites use std transcendentals, and hypot diverges more than the exp case the rule cites 2026-09-09 16:10:52 +00:00

Closed #57 A tiny non-zero drift returns a posterior variance 12 000x too small, as Ok 2026-09-09 16:10:51 +00:00

Closed #55 predict_outcome returns probabilities greater than 1: the grid clamps with no detection 2026-09-09 16:10:51 +00:00

Closed #61 Gaussian::from_ms overflows for tiny sigma, and three constructors still absorb a negative sign silently 2026-09-09 16:10:51 +00:00

Closed #56 expected_information_gain returns 4.67x its own ln k ceiling 2026-09-09 16:10:51 +00:00

Closed #60 Truncation 1 - w cancellation gives NaN skills from plausible parameters 2026-09-09 16:10:51 +00:00

Closed #59 quality() returns 0 then NaN past ~150 groups: a determinant ratio in linear space 2026-09-09 16:10:51 +00:00

Closed #58 tuple_max erases NaN, so "NaN is never convergence" is a coin flip across processes 2026-09-09 16:10:51 +00:00

Closed #38 No way to configure a competitor before their first event, and getting it wrong is silent and permanent 2026-09-08 19:15:49 +00:00

Closed #26 Test suite has no coverage for degenerate inputs, release-mode behaviour, or non-finite results 2026-09-08 19:15:46 +00:00

Closed #18 Input validation is debug-only, and the History ingestion path bypasses InferenceError entirely 2026-09-08 19:15:43 +00:00

Closed #50 ITERATIONS = 30 is far too low for a dense history, and overrunning it fails quietly 2026-09-08 19:01:23 +00:00

Closed #47 A node's reported sigma is ~10x wider than the estimate's measured reproducibility 2026-09-08 17:46:28 +00:00

Closed #5 Decide whether rayon becomes a default-on feature 2026-09-08 13:53:48 +00:00

Closed #37 EventBuilder and record_winner/record_draw cannot set prior or drift_scale 2026-09-08 13:53:48 +00:00

26 Issues created by 1 user

Opened #53 Declarative competitor configuration: HistoryBuilder::default_rating_for 2026-09-08 19:14:14 +00:00

Opened #54 Benchmark regressions are not gated — baselines are compared by hand 2026-09-08 19:14:31 +00:00

Opened #55 predict_outcome returns probabilities greater than 1: the grid clamps with no detection 2026-09-09 14:50:24 +00:00

Opened #56 expected_information_gain returns 4.67x its own ln k ceiling 2026-09-09 14:50:43 +00:00

Opened #57 A tiny non-zero drift returns a posterior variance 12 000x too small, as Ok 2026-09-09 14:51:05 +00:00

Opened #58 tuple_max erases NaN, so "NaN is never convergence" is a coin flip across processes 2026-09-09 14:51:22 +00:00

Opened #59 quality() returns 0 then NaN past ~150 groups: a determinant ratio in linear space 2026-09-09 14:51:42 +00:00

Opened #60 Truncation 1 - w cancellation gives NaN skills from plausible parameters 2026-09-09 14:52:04 +00:00

Opened #61 Gaussian::from_ms overflows for tiny sigma, and three constructors still absorb a negative sign silently 2026-09-09 14:52:28 +00:00

Opened #62 posterior_of is nondeterministic across processes: float sums over a HashMap 2026-09-09 14:52:50 +00:00

Opened #63 Three production sites use std transcendentals, and hypot diverges more than the exp case the rule cites 2026-09-09 14:53:07 +00:00

Opened #64 tests/determinism.rs never exercises the parallel sweep it exists to prove 2026-09-09 14:53:27 +00:00

Opened #65 ConstantDrift(-x) is bit-identical to ConstantDrift(x) — seal the field 2026-09-09 15:51:36 +00:00

Opened #66 log_evidence_for silently returns whole-history evidence for an unknown key 2026-09-09 17:52:35 +00:00

Opened #67 A dropped EventBuilder silently ingests nothing — no #[must_use] 2026-09-09 17:52:53 +00:00

Opened #68 The Time generic is unreachable downstream — Untimed and custom timestamps cannot be used at all 2026-09-09 17:53:12 +00:00

Opened #69 Game has no public constructor — every Game::* returns OwnedGame 2026-09-09 17:53:30 +00:00

Opened #70 Accessors disagree on how they report an unknown key, and learning_curve cannot distinguish a typo from a registered competitor 2026-09-09 17:53:50 +00:00

Opened #71 Gaussian's operators are EP internals wearing arithmetic's clothes, and / can return a confident wrong number 2026-09-09 17:54:58 +00:00

Opened #72 Ergonomics: the first program a new user writes does not compile, and the fix is not discoverable 2026-09-09 17:55:27 +00:00

Opened #73 Six exported types are unreachable from History — removing them shrinks the API for free 2026-09-09 17:55:48 +00:00

Opened #74 InferenceError: stringly-typed discriminators, and #[non_exhaustive] on the enum buys nothing while the variants lack it 2026-09-09 17:56:14 +00:00

Opened #75 Vocabulary: "rating" means three things, "sigma" means three things, and "agent" is a fourth word for competitor 2026-09-09 17:56:36 +00:00

Opened #76 History has no Debug, and trait impls are inconsistent across the public value types 2026-09-09 17:56:50 +00:00

Opened #77 Docs: no README quickstart, 112 missing_docs warnings, and the only Joint example does not compile downstream 2026-09-09 17:57:17 +00:00

Opened #78 predict_* returns Ok on a NaN-poisoned fit, and eight # Errors sections do not match the code 2026-09-09 17:57:41 +00:00