bbstats / chimeraboost

Der aufgelöste Abhängigkeitssatz konnte für diesen Bericht nicht erhoben werden: GitHub dependency-graph SBOM unavailable (404); the dependency graph may be disabled for this repository

{
  "data": {
    "repo": {
      "topics": [],
      "is_fork": false,
      "size_kb": 10054,
      "has_wiki": true,
      "homepage": null,
      "languages": {
        "Python": 1178488
      },
      "pushed_at": "2026-07-26T16:41:08Z",
      "created_at": "2026-05-26T15:05:32Z",
      "owner_type": "User",
      "updated_at": "2026-07-26T16:37:42Z",
      "description": "A Python-backend gradient boosting library",
      "is_archived": false,
      "is_disabled": false,
      "license_spdx": "Apache-2.0",
      "default_branch": "main",
      "license_spdx_raw": "Apache-2.0",
      "primary_language": "Python",
      "significant_languages": [
        "Python"
      ]
    },
    "owner": {
      "blog": null,
      "name": "Nathan Walker",
      "type": "User",
      "login": "bbstats",
      "company": "NFL",
      "location": null,
      "followers": 13,
      "avatar_url": "https://avatars.githubusercontent.com/u/81766437?v=4",
      "created_at": "2021-04-01T15:45:44Z",
      "is_verified": null,
      "public_repos": 13,
      "account_age_days": 1942
    },
    "license": {
      "state": "standard",
      "spdx_id": "Apache-2.0",
      "raw_spdx": "Apache-2.0",
      "file_present": true,
      "scorecard_found": true,
      "profile_has_license": true
    },
    "activity": {
      "releases": [
        {
          "tag": "v0.24.0",
          "kind": "minor",
          "published_at": "2026-07-24T16:21:51Z"
        },
        {
          "tag": "v0.23.0",
          "kind": "minor",
          "published_at": "2026-07-24T01:06:38Z"
        },
        {
          "tag": "v0.22.0",
          "kind": "minor",
          "published_at": "2026-07-24T00:33:22Z"
        },
        {
          "tag": "v0.21.0",
          "kind": "minor",
          "published_at": "2026-07-22T19:57:17Z"
        },
        {
          "tag": "v0.20.0",
          "kind": "minor",
          "published_at": "2026-07-21T02:44:50Z"
        },
        {
          "tag": "v0.19.0",
          "kind": "minor",
          "published_at": "2026-07-20T23:33:14Z"
        },
        {
          "tag": "v0.18.1",
          "kind": "patch",
          "published_at": "2026-07-20T11:44:20Z"
        },
        {
          "tag": "v0.18.0",
          "kind": "minor",
          "published_at": "2026-07-18T19:00:45Z"
        },
        {
          "tag": "v0.17.0",
          "kind": "minor",
          "published_at": "2026-07-18T01:19:25Z"
        },
        {
          "tag": "v0.16.1",
          "kind": "patch",
          "published_at": "2026-07-17T20:14:37Z"
        },
        {
          "tag": "v0.16.0",
          "kind": "minor",
          "published_at": "2026-07-17T18:42:29Z"
        },
        {
          "tag": "v0.15.0",
          "kind": "minor",
          "published_at": "2026-07-16T21:01:29Z"
        },
        {
          "tag": "v0.14.2",
          "kind": "patch",
          "published_at": "2026-07-14T00:46:31Z"
        },
        {
          "tag": "v0.13.0",
          "kind": "minor",
          "published_at": "2026-06-16T01:45:21Z"
        }
      ],
      "recent_commits": [
        {
          "oid": "566573aa6c87bb8416be14b148699c26ef358423",
          "body": "Docs + TabArena chart refresh (README fix needed before the 0.25.0 PyPI release)",
          "is_bot": false,
          "headline": "Merge pull request #33 from bbstats/worktree-pareto-left-selection",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-26T16:37:21Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "8b5dd833aa9c109bfe59707e3f86ca6a312fb74f",
          "body": "\"Lightning fast, near-CatBoost quality, all in Python\" (CatBoost\ncapitalised as the product name; his words otherwise).\n\nAlso the more defensible claim of the two: on the sealed TabArena holdout\nCatBoost still leads on Elo (1348 vs 1279), and our head-to-head win over\nit is on Grinsztajn, a suite we\n[…]\netter than CatBoost\" was only true on the development\nsuite.\n\nCo-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_01QL8dKTqwRwQj3j8yJSJVw3",
          "is_bot": false,
          "headline": "README headline: Nathan's wording",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-26T16:34:00Z",
          "body_truncated": true,
          "is_coding_agent": true
        },
        {
          "oid": "3abed833f1d926dbc6954f04ffb97fe0fbc2d08e",
          "body": "Nathan's call -- two ChimeraBoost points read as clutter. Chart is back to\none point per method, identical in layout to the pre-fast-rung version.\n\nThe fast rung's measurement is not thrown away: its run, leaderboard entry\nand scripts stay (benchmarks/tabarena/run_chimera_fast.py, 1243 Elo, rank\n36/\n[…]\nhe row. Nothing needs re-running if we ever want it charted.\n\nCo-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_01QL8dKTqwRwQj3j8yJSJVw3",
          "is_bot": false,
          "headline": "TabArena chart: back to the single default point",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-26T16:12:45Z",
          "body_truncated": true,
          "is_coding_agent": true
        },
        {
          "oid": "0853de55b01e2c41db288ebbfbf198eebb1b1378",
          "body": "…ethod\n\nNathan asked to see the non-ensembled rungs on the TabArena chart. Only\nTWO of the three are distinguishable there, and that was worth checking\nbefore spending the compute: TabArena runs bagged and hands every child\nan explicit eval_set, and refit_full is a no-op whenever one is supplied,\nso\n[…]\nendered as spaces.\n\nReport-only, per the sealed-holdout vow.\n\nCo-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_01QL8dKTqwRwQj3j8yJSJVw3",
          "is_bot": false,
          "headline": "TabArena-Lite: add the fast rung (quality=1) as its own leaderboard m…",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-26T15:04:49Z",
          "body_truncated": true,
          "is_coding_agent": true
        },
        {
          "oid": "2ef57f48b784b801ed15b77b337ef0c8ba7bbdec",
          "body": "Fresh 51-task Lite re-read (51/51 fitted) against the current code: A1\nvector-leaf multiclass plus the quality ladder and refit_full-by-default.\nWhole pool re-read from the regenerated leaderboard, since Elo is\nrelative.\n\nReport-only, per the sealed-holdout vow; aggregate Elo and rank only.\n\n  Chime\n[…]\narena_out/chimera_pre_select_0250 for the\nbefore/after diff.\n\nCo-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_01QL8dKTqwRwQj3j8yJSJVw3",
          "is_bot": false,
          "headline": "Refresh TabArena-Lite chart on the 0.25.0 line (Elo 1279, rank 31/68)",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-26T05:15:03Z",
          "body_truncated": true,
          "is_coding_agent": true
        },
        {
          "oid": "278cfdd6dd71f97abbdf8c2007bed05ad530b60c",
          "body": "images/pareto.png was already regenerated from the ladder co-run, but two\nplaces still transcribed the 2026-07-18 five-arm run (default 55.7% @\n5.1x) -- stale even before the default flip:\n\n- docs/PROJECT_STATUS.md \"Current standing\" now carries the full ladder\n  from results/20260725-213827.json, m\n[…]\ned in 0.24.0; it was executed to\nconfirm it runs as written.\n\nCo-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_01QL8dKTqwRwQj3j8yJSJVw3",
          "is_bot": false,
          "headline": "Propagate the new default through the Pareto text surfaces",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-26T04:34:11Z",
          "body_truncated": true,
          "is_coding_agent": true
        },
        {
          "oid": "716d235d0ab8c244ae7f9a62739bdd58821d1832",
          "body": "quality=1..5 ladder + refit_full default: ChimeraBoost owns the whole Pareto frontier",
          "is_bot": false,
          "headline": "Merge pull request #32 from bbstats/worktree-pareto-left-selection",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-26T04:29:05Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "00fef2f26c6655c5189ffb6c19cd59da1c35ed99",
          "body": "Nathan's call: the default should be whichever rung has the best accuracy\nwithout ensembling. That is rung 3 -- 69.9% of head-to-head matchups won\nagainst rung 2's 45.6%.\n\nRe-charted so the arm labelled ChimeraBoost IS the default\n(results/20260725-213827.json): Ens8 94.0% @ 25.66x, Ens5 83.7% @ 17.\n[…]\nt changes every ordinary fit -- a pre-existing coverage gap.\n\nCo-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_01QL8dKTqwRwQj3j8yJSJVw3",
          "is_bot": false,
          "headline": "Default to rung 3: refit_full=True, the strongest non-bagged setting",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-26T04:19:34Z",
          "body_truncated": true,
          "is_coding_agent": true
        },
        {
          "oid": "5136b19096fda66b88248154247313bed24374f8",
          "body": "Ladder co-run (results/20260725-155127.json, Grinsztajn, 3 seeds, every\nrung in ONE run so the column is comparable) puts all five rungs on the\nPareto frontier: 31.1% @ 1.95x, 45.6% @ 5.22x, 69.9% @ 9.15x, 83.7% @\n17.07x, 94.0% @ 24.74x. LightGBM holds only the extreme-left corner at\n21.6% @ 1.14x; \n[…]\nand winrate_matrix.png from the ladder run.\n599 tests green.\n\nCo-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_01QL8dKTqwRwQj3j8yJSJVw3",
          "is_bot": false,
          "headline": "SELECT closed: five-rung ladder charted, quality shipped, docs",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-25T20:21:15Z",
          "body_truncated": true,
          "is_coding_agent": true
        },
        {
          "oid": "2d1c66d7aa9a0199e42265a386a82ed89c8853fa",
          "body": "A user-facing ladder over the frontier the SELECT program measured. Each\nrung only pins parameters that already exist -- nothing new is computed\n-- so quality=None (the default) leaves every path byte-identical, and\nquality=2 is the shipped defaults exactly. Both verified by test.\n\n  1 fast      lin\n[…]\n69 existing tests green (goldens included) plus 30 new ones.\n\nCo-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_01QL8dKTqwRwQj3j8yJSJVw3",
          "is_bot": false,
          "headline": "quality=1..5: named operating points on the strength/slowdown Pareto",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-25T19:51:40Z",
          "body_truncated": true,
          "is_coding_agent": true
        },
        {
          "oid": "cf796dfdb7bf7cbf1614c0fe34da9edc9ddacb55",
          "body": "…y bar\n\nGrinsztajn, 3 seeds, 57 datasets scored head-to-head, all arms in one run\nagainst an identical field (no Ens8, as registered -- so absolute win\nrates are not comparable to the charted 55.7%).\n\nChimeraBoostOneLin -- one booster fit, no selection, linear leaves pinned\non -- reaches 52.0% win r\n[…]\nent); the bars are pure inequalities, so no verdict\nchanged.\n\nCo-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_01QL8dKTqwRwQj3j8yJSJVw3",
          "is_bot": false,
          "headline": "SELECT Phase 0: the 1x-5x Pareto gap is reachable; OneLin clears ever…",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-25T19:20:48Z",
          "body_truncated": true,
          "is_coding_agent": true
        },
        {
          "oid": "a4bd419dbfb25140522565537cf3c78f671fda9f",
          "body": "Reuses make_pareto.score_models so the win rates and slowdowns the bars\nare judged on are exactly the ones the chart would draw, and reports the\nfrontier membership alongside. Bar C is deliberately not reimplemented --\ncompare_runs.py already does paired sign tests with the near-solved\nguard, and ac\n[…]\nth --model / --model-new to\ncompare two arms within one run.\n\nCo-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_01QL8dKTqwRwQj3j8yJSJVw3",
          "is_bot": false,
          "headline": "select_read: chord test for the SELECT bars",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-25T19:08:24Z",
          "body_truncated": true,
          "is_coding_agent": true
        },
        {
          "oid": "b57e99ac510ec28991a1e7cd6dedb2fa14864022",
          "body": "…arms\n\nThe charted Pareto has nothing between LightGBM (1.0x) and ChimeraBoost\n(4.9x). The cost structure suggests why that gap should be reachable: a\ndefault regression fit runs up to four boosters (two capped auditions, a\ncross-augmented candidate, and a from-scratch refit of the winner when\ncross\n[…]\nday kill clause.\n\nBenchmarks-only; the library is untouched.\n\nCo-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_01QL8dKTqwRwQj3j8yJSJVw3",
          "is_bot": false,
          "headline": "SELECT_PLAN: pre-registered selection-economics probe + fixed-config …",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-25T19:07:25Z",
          "body_truncated": true,
          "is_coding_agent": true
        },
        {
          "oid": "0ba0f109ffc182511307bfaed934ee85be65b5d8",
          "body": "Fix compare_runs' near-solved mean (+ A2 Phase 0 record)",
          "is_bot": false,
          "headline": "Merge pull request #31 from bbstats/worktree-a2-config-portfolio",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-25T18:54:19Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "44a46c4a7fa579f8b3a8ed5766284a13cab5ffd2",
          "body": "compare_runs is the tool every /experiment ship decision runs through, and\nthe ship rule is \"decisive sign test + MEAN improvement on Grinsztajn AND a\nnon-negative OpenML gate\" -- so the mean is load-bearing. It had no\nnear-solved guard at all, while summarize.py and make_tables.py both do.\n\nDemonst\n[…]\nsts/test_compare_runs.py pins the contracts. 569 tests pass.\n\nCo-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_018u2u5K96eQ8aamSM5j6tsu",
          "is_bot": false,
          "headline": "Fix compare_runs' mean: exclude near-solved datasets, report median",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-25T17:25:26Z",
          "body_truncated": true,
          "is_coding_agent": true
        },
        {
          "oid": "e180960f4d753be76a49b475b7bce6e74a571b28",
          "body": "156 fits (13 PMLB tune sets x 3 seeds x 4 configs), no library change.\n\nMeasurement guard: pm:tune/nursery is solved to numerical zero (Brier\n~1e-49 for every config) and manufactured a -8e21% mean delta on the first\npass. Dropped by summarize.py's own near-solved convention, leaving 36 of\n39 cells.\n[…]\nn, high-card, the OpenML gate and TabArena were not touched.\n\nCo-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_018u2u5K96eQ8aamSM5j6tsu",
          "is_bot": false,
          "headline": "A2 Phase 0: pre-registered portfolio KILLED on cost; depth race survives",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-25T17:05:09Z",
          "body_truncated": true,
          "is_coding_agent": true
        },
        {
          "oid": "1a751112fd34835763b2e49671f81ca772f5a4d2",
          "body": "Pre-registration before any result was read: design, portfolio, kill bars\nA-D, and stated predictions. Phase 0 measures per-dataset config headroom\nand simulates the capped race offline from recorded validation curves.\n\nProtocol call: Phase 0 runs on the PMLB TUNING fold only. A2 is a\nhyperparameter\n[…]\necision suites, the OpenML gate and TabArena stay untouched.\n\nCo-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_018u2u5K96eQ8aamSM5j6tsu",
          "is_bot": false,
          "headline": "A2_PLAN: pre-registered config-portfolio audition + Phase 0 tooling",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-25T16:58:00Z",
          "body_truncated": true,
          "is_coding_agent": true
        },
        {
          "oid": "b4cff091d285dd1ef1ee097647b5186f6182f785",
          "body": "A1: vector-leaf multiclass with sketched split scoring",
          "is_bot": false,
          "headline": "Merge pull request #30 from bbstats/a1-vector-leaf",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-24T19:22:06Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "bc78349c7d555a19c994d31a6f0b9a7e9d511d59",
          "body": "…dict\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_016xhoePdEMTMgJkSh93CzP6",
          "is_bot": false,
          "headline": "A1 SHIP: gate 10W-0L, goldens re-blessed, CHANGELOG + docs + plan ver…",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-24T19:09:10Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "b55756f55a45b368a44373254eb319a85657a380",
          "body": "Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_016xhoePdEMTMgJkSh93CzP6",
          "is_bot": false,
          "headline": "A1 hc 6-seed extension PASS on pre-stated instrument",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-24T18:59:12Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "83b2e057c860cb126bfb71ed1641643a6a648036",
          "body": "Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_016xhoePdEMTMgJkSh93CzP6",
          "is_bot": false,
          "headline": "A1 tier-2 hc read + pre-stated 6-seed extension on the 4 mc sets",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-24T18:56:18Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "346e0dd3ef75e933fd927fc1eb0948edd46f518e",
          "body": "…%, fit 2.5x faster\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_016xhoePdEMTMgJkSh93CzP6",
          "is_bot": false,
          "headline": "A1 tier-1 screen PASS: identity 102/102, F1 parity, Brier 23W-8L +5.0…",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-24T18:39:58Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "109e99e2bbad1322c1d6efd192d7a82c0474ba1f",
          "body": "…oring\n\nOne tree per round with K-vector Newton leaves replaces K trees per round.\nSplit search runs on a 1-d centered Rademacher projection of the gradient\ncolumns (scalar split kernels, quantized path included, reused verbatim);\nprojected curvature hess@r^2 keeps min_child_weight/l2 semantics stab\n[…]\nkles. Goldens for multiclass panel sets re-bless at\nship per A1_PLAN.md.\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_016xhoePdEMTMgJkSh93CzP6",
          "is_bot": false,
          "headline": "A1: vector-leaf multiclass with centered-Rademacher sketched split sc…",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-24T18:36:33Z",
          "body_truncated": true,
          "is_coding_agent": true
        },
        {
          "oid": "a7340872235bccb3d82e6e8bfa0d149e771a845d",
          "body": "…multiclass\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_016xhoePdEMTMgJkSh93CzP6",
          "is_bot": false,
          "headline": "A1_PLAN: pre-registered design/predictions/kill bars for vector-leaf …",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-24T18:24:35Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "fe859f44242242d5559f2292e1ae1bb3e546d3bd",
          "body": "refit_full: opt-in full-data refit of the ES winner (all /experiment gates PASS)",
          "is_bot": false,
          "headline": "Merge pull request #29 from bbstats/lre-cheap-ensemble",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-24T17:39:21Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "b3474fa690b530d3a817b293eec3517d6c4b6b6c",
          "body": "… verdict\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_01Jf5H6AVHQx1FeMHjECDCFy",
          "is_bot": false,
          "headline": "refit_full stays opt-in per Nathan: docs (parameters + recipes), plan…",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-24T17:37:19Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "a0a7abc0ee3dd162fa1870486c77ec5925059ec0",
          "body": "Tier 1 synth 95W-39L +1.02%; Grinsztajn 48W-11L +2.00% (Brier 18W-5L\n+2.43%); hc 10W-3L-1T +0.57% (Brier 8W-0L); OpenML one-shot 26W-8L-2T\n+1.27%. Spliced headline: single win rate 55.7->73.7% (external field\n93.0%), CatBoost 50.9->43.4, fit x1.98. Default stays False pending\nspeed sign-off.\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_01Jf5H6AVHQx1FeMHjECDCFy",
          "is_bot": false,
          "headline": "REFIT_PLAN results: all gates PASS; CHANGELOG [Unreleased] refit_full",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-24T17:14:40Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "619bd19946a34003fd8cdb180847c47dad43d38c",
          "body": "…t probes\n\n- refit_full param on both estimators: after the auto ES split has chosen\n  the budget (selection/calibration done), retrain the winner on 100% of\n  rows, rounds scaled by the train-size ratio, lr pinned; temperature/CQR\n  transfer; explicit eval_set / early_stopping=False / quantile / ba\n[…]\ntructural diversity).\n- --chimera-refit-full harness flag; 10 new tests.\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_01Jf5H6AVHQx1FeMHjECDCFy",
          "is_bot": false,
          "headline": "refit_full: full-data refit of the ES winner (default off) + LRE/refi…",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-24T16:53:41Z",
          "body_truncated": true,
          "is_coding_agent": true
        },
        {
          "oid": "52d0aae8c645589783c2f6cb5c99860091348f21",
          "body": "…Poisson/Gamma/Tweedie)\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_01Jf5H6AVHQx1FeMHjECDCFy",
          "is_bot": false,
          "headline": "Release 0.24.0 (custom objectives + eval_metric hook; built-in Huber/…",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-24T16:17:31Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "377deb80ab97e315bccd44b8072fdb4740624f38",
          "body": "Custom objectives + eval_metric hook; built-in Huber/Poisson/Gamma/Tweedie",
          "is_bot": false,
          "headline": "Merge pull request #28 from bbstats/worktree-custom-objectives",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-24T16:14:34Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "60cc982fff596a6f09dee4f2fd307a7b195351f1",
          "body": "…eedie\n\nPhase 3 backlog top pick (custom objective + eval_metric hook, which\nunlocks the classic extra regression losses for free):\n\n- Regressor loss accepts a user objective instance (subclass\n  chimeraboost.CustomObjective: grad_hess + eval, optional init/transform).\n  Non-string losses bypass the\n[…]\n-identity goldens and a custom-objective-vs-builtin exact-equality\ntest.\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_01Jf5H6AVHQx1FeMHjECDCFy",
          "is_bot": false,
          "headline": "Custom objectives + eval_metric hook; built-in Huber/Poisson/Gamma/Tw…",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-24T01:41:14Z",
          "body_truncated": true,
          "is_coding_agent": true
        },
        {
          "oid": "4e36a660684faa7567edcf6e41e1c8fc4a16a4ab",
          "body": "…tical, ChimB 1278 rank 31/68; predict 0.162->0.125 s/1K)\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_01WD2LMMYqx1nyxxE7pKRcUv",
          "is_bot": false,
          "headline": "Refresh TabArena pareto (post-0.23.0 Lite re-read: Elo pool byte-iden…",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-24T01:22:50Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "6667843b8970454b0f582ffd1ab2be033989c578",
          "body": "…plit, serial predict twins, cross prealloc)\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_01WD2LMMYqx1nyxxE7pKRcUv",
          "is_bot": false,
          "headline": "Release 0.23.0 (PR #27 follow-ups: pair-code combos, seeded grouped s…",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-24T01:05:31Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "8dbe852fa39c1a03127cec488f556c77aa8e3370",
          "body": "…hanged, ChimB 4.9x / Ens8 23.6x)\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_01RSMP6TRSv1WzFBjcRnvTJs",
          "is_bot": false,
          "headline": "Refresh pareto charts (post-0.22.0 run 20260723-192007; win rates unc…",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-24T00:32:20Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "d25da33d353d3f939fb428d29f0ca1f2f2afcb2a",
          "body": "Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_01RSMP6TRSv1WzFBjcRnvTJs",
          "is_bot": false,
          "headline": "Release 0.22.0 (version bump; artifact built at 6bbda8c, tagged v0.22.0)",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-24T00:32:20Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "5fd76b8683dd49bc1ee829cfcd26ec48022a4c88",
          "body": "Audit backlog: pair-code combos, grouped-split random_state, serial predict twins",
          "is_bot": false,
          "headline": "Merge pull request #27 from bbstats/backlog-semibug-followups",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-24T00:30:57Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "979b62a86bb1d783c63630b12e518e68f0e36664",
          "body": "…ial predict twins, cross prealloc\n\nRemaining items from the 2026-07-23 semi-bug audit (525 tests incl.\nidentity goldens green):\n\n- Combo categoricals are (val_a, val_b) pairs, not \"_x_\" strings: fixes\n  delimiter-collision aliasing, NaN-vs-\"nan\" conflation, int/float\n  spelling splits. Identical gr\n[…]\ny.\n- Bool columns no longer misnamed in the cat_features error guidance.\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_01RSMP6TRSv1WzFBjcRnvTJs",
          "is_bot": false,
          "headline": "Backlog follow-ups: pair-code combos, grouped-split random_state, ser…",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-23T23:42:35Z",
          "body_truncated": true,
          "is_coding_agent": true
        },
        {
          "oid": "7951d6d717ca2371e240045133e6546e934efe8c",
          "body": "Predict/fit semi-bug fixes: cross-block cast reuse, single predict conversion, MAE lei guard, depth-0 truncation",
          "is_bot": false,
          "headline": "Merge pull request #26 from bbstats/predict-fit-semibug-fixes",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-23T23:05:14Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "2a8be66e3833d4837109c1ff4e6ff0141384db3a",
          "body": "…, MAE lei guard, depth-0 truncation\n\nAudit follow-up to the 0.21.0 categorical-predict findings. All changes are\nbit-identical on default paths (519 tests incl. identity goldens green).\n\nPerf (bit-identical):\n- _cross_block reads numeric parents from the transform's numeric block:\n  one float64 cas\n[…]\nt validation prefix like every\n  other exit (scalar + multiclass loops).\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_01RSMP6TRSv1WzFBjcRnvTJs",
          "is_bot": false,
          "headline": "Predict/fit semi-bug fixes: cross-block cast reuse, single conversion…",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-23T22:57:46Z",
          "body_truncated": true,
          "is_coding_agent": true
        },
        {
          "oid": "26fed8a715fe172518472f4fec1a663492db6f61",
          "body": "Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_018V1LjsaDPtHhjvMCbAKkKw",
          "is_bot": false,
          "headline": "Release 0.21.0",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-22T19:56:07Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "b2139dca2215ffd01af3d9199af0cc1ca2c69dd0",
          "body": "Drop pandas dependency; share categorical transform across bagged members at predict",
          "is_bot": false,
          "headline": "Merge pull request #25 from bbstats/bagged-cat-predict",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-22T19:53:16Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "827d67f12c4339583d6a9fb426d407b92e072ebb",
          "body": "…predict\n\nThe categorical machinery is now numpy/numba only, bit-identical to the\npandas implementation (the gdiff group sums replicate pandas' Kahan-\ncompensated groupby exactly; verified bitwise across reg/binary/multiclass\nx single/bagged x weighted x DataFrame/ndarray, incl. fit matrices and\ngdi\n[…]\n 513 + 4 new\ntests green, incl. a pandas-import-blocked end-to-end test.\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_018V1LjsaDPtHhjvMCbAKkKw",
          "is_bot": false,
          "headline": "Drop pandas dependency; share cat transform across bagged members at …",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-22T19:43:53Z",
          "body_truncated": true,
          "is_coding_agent": true
        },
        {
          "oid": "852481bb9cc99c03ac182a4feb00274784eadbed",
          "body": "…pped)\n\nA parallel n_ensembles fit divides the thread budget across workers, but\nmembers kept the budget/K sliver permanently and predict sequentially --\nan 8-member bag walked each forest on 1-2 threads. Bit-identical\npredictions; bagged predict ~3-5x faster (50k rows, K=8: numeric\n0.80->0.24s, categorical 2.31->1.60s).\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>",
          "is_bot": false,
          "headline": "Bagged members: restore parent thread_count after fit (predict was ca…",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-22T19:02:07Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "6a76586dfdff90275e7e816f25e35c927b8527fb",
          "body": "Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_01AphtGuN8Uxr74JZKamikH8",
          "is_bot": false,
          "headline": "Pareto charts: linear x-axis (log was overkill; same dimensions)",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-21T02:47:22Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "0a3163a0b33b8933554432d499fc69ffa10025ae",
          "body": "…, in-CI), train 0.83 predict 0.162 s/1K\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_01AphtGuN8Uxr74JZKamikH8",
          "is_bot": false,
          "headline": "TabArena chart: post-0.20.0 Lite re-read — Elo 1264->1278 (rank 31/68…",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-21T02:39:42Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "4e1b4e25ab0fd7f75e4f33241bf5f5925f5fda9e",
          "body": "… Ens8 23.9x->23.2x\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_01AphtGuN8Uxr74JZKamikH8",
          "is_bot": false,
          "headline": "Refresh pareto: 0.20.0 field — win rates unchanged, ChimB 5.1x->4.8x,…",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-21T02:25:56Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "7d48e053e5bd3c7aded1126871aeb0f1f6b84c46",
          "body": "…oss skip\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_01AphtGuN8Uxr74JZKamikH8",
          "is_bot": false,
          "headline": "Release 0.20.0: group-centered categorical crosses + audition train-l…",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-21T02:07:00Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "8c6f1499f90fa8ed07ac2badd892562442aa4bb5",
          "body": "Cross features for categorical columns + faster selection fits",
          "is_bot": false,
          "headline": "Merge pull request #24 from bbstats/worktree-pareto-catcross",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-21T01:56:59Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "301e2f88ad94e429f6d3d3385848a5ef620a5b02",
          "body": "Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_01AphtGuN8Uxr74JZKamikH8",
          "is_bot": false,
          "headline": "CATCROSS: T3 gate flat (non-negative) — SHIP verdict; CHANGELOG + docs",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-21T01:39:45Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "e5ece0324afb0b7bae68aa135b0e587ad772dfb9",
          "body": "…wup into +2.53%)\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_01AphtGuN8Uxr74JZKamikH8",
          "is_bot": false,
          "headline": "CATCROSS: T2 highcard PASS (5W-3L +0.254%, race flipped the probe blo…",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-21T01:23:32Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "2040e3c1d758ad00dd08cdc556719f71fb0fbf03",
          "body": "…ver passed)\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_01AphtGuN8Uxr74JZKamikH8",
          "is_bot": false,
          "headline": "CATCROSS: T2 Grinsztajn 59/59 exact ties (structurally inert, cats ne…",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-21T01:09:22Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "40542a8a41ec8c3fdac3f286f34881fc0e317963",
          "body": "Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_01AphtGuN8Uxr74JZKamikH8",
          "is_bot": false,
          "headline": "CATCROSS: T1 synth PASS (entity-slice concentration, controls exact)",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-21T00:49:40Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "0b3c312f8377ec88cdbb6bf2a54b923a86a424fe",
          "body": "Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_01AphtGuN8Uxr74JZKamikH8",
          "is_bot": false,
          "headline": "CATCROSS_PLAN: pre-register gdiff predictions, kill clauses, ship bar",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-21T00:00:18Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "e0d401bab9b16041f0323ae923e43dfa413532c3",
          "body": "New cross-pair op 'gdiff': a column holding each row's numeric value minus\nits category's fit-time mean, so 'above this row's own category's baseline'\nbecomes one split instead of a per-category staircase (the num x cat analog\nof the 2026-07-13 diff/prod fix). Target-free -- the same fitted map serv\n[…]\nrees. Decision suites pending per /experiment.\n\n513 tests green (9 new).\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_01AphtGuN8Uxr74JZKamikH8",
          "is_bot": false,
          "headline": "Group-centered categorical crosses (gdiff): x_num - mean(x_num | cat)",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-20T23:58:27Z",
          "body_truncated": true,
          "is_coding_agent": true
        },
        {
          "oid": "a01de47538f079180c84e98e17b684c3dcbf94e5",
          "body": "The raced-selection auditions install internal callbacks that never read\ntrain_loss, but the fit loop evaluated the loss on the full training set\nevery round whenever any callback was present. Internal callbacks are now\ntagged _cb_needs_train_loss=False and the loop passes None instead of\npaying the\n[…]\nault\nbinary fit 15.6% faster on a 50k-row A/B (0.959s -> 0.809s median).\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_01AphtGuN8Uxr74JZKamikH8",
          "is_bot": false,
          "headline": "Skip per-round train-loss eval for selection-audition callbacks",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-20T23:48:10Z",
          "body_truncated": true,
          "is_coding_agent": true
        },
        {
          "oid": "919a80a65cc77657aa4c9f8d100227bb61377d84",
          "body": "A1 + 0.19.0 release prep: verbose notice for the silent early-stopping holdout",
          "is_bot": false,
          "headline": "Merge pull request #23 from bbstats/worktree-a1-es-notice",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-20T23:27:08Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "a564bbfc756ce8e774ae1aa21fdad18c7509153a",
          "body": "…g holdout\n\nWith verbose=True, fit now prints how many rows early_stopping=True held\nout as the auto validation set (both estimators; bag members stay quiet).\nDefault verbose=False unchanged. Closes the last decided Phase 2 audit\nitem. Bundles the 0.19.0 version bump + CHANGELOG (H1/H2 lei auto\ndefault, H3 weight-aware fixes, this notice).\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_01J3G7QxkKLXfh3DinQiMqqk",
          "is_bot": false,
          "headline": "A1 + release prep 0.19.0: verbose notice for the silent early-stoppin…",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-20T23:20:49Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "266f899b657a3b875c5bc1ee02d5ba8c593ad459",
          "body": "H3: make sample_weight weight-aware (encoder, binner, early-stopping metric)",
          "is_bot": false,
          "headline": "Merge pull request #22 from bbstats/worktree-h3-weight-aware",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-20T17:07:30Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "607e70a39397b85cb63b7ad25b234ab5d0580b2d",
          "body": null,
          "is_bot": true,
          "headline": "Merge main: resolve CHANGELOG conflict (H1/H2 + H3 Unreleased entries)",
          "author_name": "copilot-swe-agent[bot]",
          "author_login": "Copilot",
          "committed_at": "2026-07-20T16:51:49Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "0d9804e481f7ff0ee66615496ed9e5b54772b169",
          "body": "sample_weight was only applied to the gradient/leaf step. Rows with\nweight 0 still shaped the ordered-target categorical encodings, the\nquantile bin borders, and — most damaging — the early-stopping metric,\nwhich was computed unweighted on the auto-split validation fold. In an\nadversarial repro (zer\n[…]\nion (they use no weights), confirmed by the\nnumerical-identity goldens.\n\nCo-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_01C73oLdSQaovFVQzxRkDQzg",
          "is_bot": false,
          "headline": "H3: make sample_weight weight-aware in encoder, binner, and ES metric",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-20T16:40:47Z",
          "body_truncated": true,
          "is_coding_agent": true
        },
        {
          "oid": "0be8d109d6d6c664870ddec959d32e5356cb624c",
          "body": "H1/H2: honest leaf_estimation_iterations default (3 → None auto)",
          "is_bot": false,
          "headline": "Merge pull request #21 from bbstats/worktree-h1h2-lei",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-20T14:41:57Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "a9237cd6dd4fc94970652941590165762a33e47d",
          "body": "…, honest warnings\n\nPhase-2 audit follow-up. An A/B across synth, Grinsztajn, highcard and the\nOpenML gate KILLED the hypothesis that clf lei=3 was a stale default:\n- Grinsztajn 59/59 bit-identical (linear leaves fully shadow lei there),\n  reg 0-0-48 and multiclass 0-0-34 exact ties on synth, highca\n[…]\nd default and needs its own gate; recorded as a follow-up.\n\nCo-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_01DH3g2jgrBetEkUib3xV5By",
          "is_bot": false,
          "headline": "H1/H2: classifier leaf_estimation_iterations default 3 -> None (auto)…",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-20T14:36:39Z",
          "body_truncated": true,
          "is_coding_agent": true
        },
        {
          "oid": "1b58aa26d1ef2487398723166399937e9260ae72",
          "body": "Audit Phase 1 fixes (thread hygiene, pickle slim, loud failures, docs)\nplus the H5 early-stop truncation fix.\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_01Bpqad7wW42HtzYJz633fwq",
          "is_bot": false,
          "headline": "Release 0.18.1",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-20T11:41:39Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "9a5a3fd76088eb68244a7cc172d216aeec2091ce",
          "body": "Exhausting n_estimators before patience fired kept every tree built past\nthe best round, contradicting the best_iteration_ contract (~5% val RMSE\ncost in the budget-exhausted repro). for/else keeps callback stops (the\nselection auditions) byte-identical; verified synth 136/136 exact ties,\nGrinsztajn\n[…]\nhe only\nexhausted fits, OpenML gate 35/36 ties + one sub-0.005% epsilon.\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_01Bpqad7wW42HtzYJz633fwq",
          "is_bot": false,
          "headline": "H5 fix: truncate to best iteration when the tree budget beats patience",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-20T01:35:32Z",
          "body_truncated": true,
          "is_coding_agent": true
        },
        {
          "oid": "f14be606b641f1bf0dc92bb14b3951f1fe631c6b",
          "body": "…ions\n\nCodebase step-back audit (5 parallel review agents + repro'd findings).\nBehavior-neutral fixes only; all fitted models bit-identical (483 tests\nincl. numerical-identity goldens green, 14 new regression tests).\n\nCode:\n- thread_count now applies to predict and the global numba setting is\n  rest\n[…]\ny)\n\nGetting-started doctest outputs verified byte-exact on current code.\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>\nClaude-Session: https://claude.ai/code/session_014FHt54zCikygcZHP3YpBnn",
          "is_bot": false,
          "headline": "Audit fixes: thread hygiene, pickle slim, loud failures, docs correct…",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-19T15:33:40Z",
          "body_truncated": true,
          "is_coding_agent": true
        },
        {
          "oid": "12dcdcda91035d61c540cf9b309916765ccafb18",
          "body": "test_strength_viz imports make_pareto, and CI installs only the library\ndeps (no matplotlib) -> collection error killed the whole suite on every\nPython version (run 29664678946). matplotlib now imports inside the two\nrender_* functions only; scoring/tables/tests need none of it. Verified\nlocally with matplotlib shadow-blocked: 8/8 pass, 469 tests collect.\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>",
          "is_bot": false,
          "headline": "Fix CI: lazy matplotlib import in make_pareto",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-19T00:00:02Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "97ca6b86b7cb9f7ba110675e350f24e7b183a6ee",
          "body": "Per STRENGTH_VIZ_PLAN (Nathan resolved D0 to win percentage). Blended-%\nsaturated at 99.x on near-Bayes-optimal data; win rate (% of dataset x\nopponent matchups won on RMSE reg / Brier clf, ties 1/2, 95% bootstrap CI)\ndecompresses the field: Ens8 98.2 / ChimB 55.7 / CatB 50.9 / LGBM 28.5 /\nHGB 16.7 \n[…]\ndocs/PROJECT_STATUS 2, /pareto skill, CLAUDE.md north-star line updated\n- Ship-gating unchanged: chart legibility only, library source untouched\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>",
          "is_bot": false,
          "headline": "Strength-viz shipped: pareto.png headline axis = head-to-head win rate",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-18T23:07:44Z",
          "body_truncated": true,
          "is_coding_agent": true
        },
        {
          "oid": "00ecb25fe5009f850e46f50a8dde4e45aeb5fd7d",
          "body": "Win rate vs the field is mean rank linearly rescaled (winrate =\n(k - mean_rank)/(k - 1), ties half), so the intuitive unit costs nothing.\nOutright-winner share explicitly rejected as the axis (degenerate with\nEns8 in-pool). D1-D3 remain open for the build session.\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>",
          "is_bot": false,
          "headline": "Strength-viz plan: axis resolved to pairwise win rate per Nathan",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-18T22:37:07Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "d525d335816c1c79b050765636fd003c661ae5a7",
          "body": "…bility)\n\nPre-registered plan (no code change). Blended percent-vs-best saturates near\n100 on near-optimal tabular data, so 4 of 5 models smear into a 2-point band.\nRecommends: headline y-axis to mean rank (bootstrap CIs) plus a head-to-head\nwin-rate matrix companion; keep blended as a diagnostic. Per-dataset scores\nalready live in the result JSONs, so it is pure re-analysis with no new runs.\nThree decisions flagged for the build session.\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>",
          "is_bot": false,
          "headline": "Plan: make the Grinsztajn chart show winner/loser (strength-axis legi…",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-18T22:35:04Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "2044329f91192526d4dbb30b611dfcbf2193e217",
          "body": "…flat\n\n51-task Lite re-run (0/51 fail) after quantize_gradients shipped default-on\nin 0.18.0. ChimeraBoost Elo 1267 -> 1264 (-3, inside +-58/60 CI = flat),\nrank 32/68 unchanged; accuracy held after the default flip. Pool re-read\nsame-run (CatBoost 1346->1348, XGB/LGBM/RF/Linear exact). Train 0.79->0\n[…]\n 20-26% fit win\nquantization gives on Grinsztajn/high-card (known Lite property). Still on\nthe frontier. Report-only per the sealed-holdout vow.\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>",
          "is_bot": false,
          "headline": "TabArena-Lite chart refresh: 0.18.0 (quantized histograms) Elo 1264, …",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-18T22:26:03Z",
          "body_truncated": true,
          "is_coding_agent": true
        },
        {
          "oid": "dc74afabe4e22af96d499409b1f9dfa68bdceced",
          "body": "Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>",
          "is_bot": false,
          "headline": "Release 0.18.0",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-18T18:54:32Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "dfd040011b008ec0804071b8791f855eaa058946",
          "body": "Fresh canonical 5-arm gr run 20260718-142950: accuracy columns identical\nto 20260717-153114 to the decimal, competitors unmoved, both Chimera\npoints moved left. Program record closed in QUANT_PLAN.md.\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>",
          "is_bot": false,
          "headline": "Q-program close: pareto refresh (single 6.4x->5.1x, Ens8 31.0x->23.9x)",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-18T18:47:26Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "a321e60b1831284890dc61fb137851283eac2fdb",
          "body": "All registered Phase-3 bars passed (full record benchmarks/QUANT_PLAN.md):\ntier-1 synth flat (28W-33L-75T, +0.006%, canaries exactly 0; Brier flat);\ngr accuracy parity (27W-24L-8T, -0.036%) with fit sum 0.738 (-26.2%);\nhc parity both arms (3-3-8 / Ens8 6-5-3) with fit sums 0.796 / 0.805;\nOpenML gate\n[…]\nrol 0.989.\nGoldens regenerated (behavior change, cross_features precedent);\n461 tests green; quantize_gradients=False restores exact float path.\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>",
          "is_bot": false,
          "headline": "Q-program SHIP: quantize_gradients default ON; goldens, CHANGELOG, docs",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-18T18:29:19Z",
          "body_truncated": true,
          "is_coding_agent": true
        },
        {
          "oid": "eeba9ef327f119799c7429d5f9fc05182bd44590",
          "body": "Harness A/B mechanics for QUANT_PLAN Phase 3: --chimera-quantize sets\nquantize_gradients=True on every ChimeraBoost arm (single + bagged\nmembers), stamped into the run config line. quant_smoke.py: real\nwrapper-level fits engage the flag end to end — reg 1.22-1.23x fit,\nbin 1.06-1.09x (ridge/calibration untouched by design), metrics ~identical.\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>",
          "is_bot": false,
          "headline": "Q-program: --chimera-quantize harness flag + wrapper-level smoke",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-18T18:13:04Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "1fa8120f3ccdb98f9f6acd65d4877ec3c2ce31c9",
          "body": "…a flag\n\nQUANT_PLAN.md program (Nathan's FP-drift opt-in, 2026-07-18). Split search\nruns on ~15-bit quantized grad/hess packed one-int64-per-sample (LightGBM-4\n-style): one integer RMW per scatter write, half the histogram footprint.\nLeaf values / ridge / LOO keep exact float grads (leaf renewal for\n[…]\nefault path bit-identical, goldens untouched. Oracle test: bit-exact vs\nfloat kernel on power-of-two-representable data, both dispatch branches.\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>",
          "is_bot": false,
          "headline": "Q-program Phase 1: quantized-gradient packed-int64 histograms behind …",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-18T18:08:35Z",
          "body_truncated": true,
          "is_coding_agent": true
        },
        {
          "oid": "9020b414878d719b442f0eb62e9a7c5ecdbf1b30",
          "body": "L-pytree ships as bit-identical engineering (73/73 + Ens8 identity ties;\n~1% suite sums, real small-fit wins). Program closed acceptance-not-met;\nfull record in benchmarks/GROW_PLAN.md.\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>",
          "is_bot": false,
          "headline": "Merge grow-kernels: fused level kernel + grow attribution program record",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-18T17:27:06Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "011ac3305882d5c27cf0bd9550f56f1e4fb4dd91",
          "body": "Tier-2 identity PASS in full (gr 59/59, hc 14/14 both arms); fit-time\nsums gr 0.989 / hc 1.018 / Ens8 0.984 — below the registered 10% bar,\nkill clause engaged. L-pytree retained (unreleased CHANGELOG entry);\nPhase 2 not pursued per registered default. PARETO_PLAN updated.\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>",
          "is_bot": false,
          "headline": "Grow-kernel program CLOSED: acceptance not met, verdicts recorded",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-18T17:24:25Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "3fd741c46fe306d565a328623e3de575918250da",
          "body": "pytree residue collapsed on the whole panel (15.4->5.3% worst set),\nms/tree -3..13% on gr; 10/10 exact prediction fingerprints vs main.\nTier-2 identity runs in flight.\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>",
          "is_bot": false,
          "headline": "L-pytree timing read + clean-box smoke PASS recorded; grow_smoke.py",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-18T17:15:22Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "12d0f74c73014b94b5b820cedf70ea7a50f08864",
          "body": "_build_split_descend = _build_and_split search + in-kernel descend +\nnext-level occupancy list (small n), one launch per level instead of a\nsplit launch + descend launch + bincount/flatnonzero pair. Old kernels\nretained as oracles; exact-equality pipeline test added; warmup updated\n(one signature covers both branches); 455 tests incl. identity goldens\ngreen. Timing read to follow.\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>",
          "is_bot": false,
          "headline": "L-pytree: single-launch level kernel, bit-identical (GROW_PLAN Phase 1)",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-18T16:46:44Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "5965bd516fec526021f52349fd5fff5c42d0bcf0",
          "body": "Kernel-vs-kernel micro: x0.49-0.67 at n>=37.5K — the ridge is\nFMA-bound, not gather-bound. Ridge-micro mode guarded; numba\n__wrapped__ gotcha recorded.\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>",
          "is_bot": false,
          "headline": "L-ridge verdict: KILLED (measured regression, reverted); record why",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-18T16:41:58Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "686219eb80194da960ae392b3b78cacbfbf13d5f",
          "body": "…AN Phase 1)\"\n\nThis reverts commit 332fd39839f02b9642a12bf6d4f5e0d657ff0e51.",
          "is_bot": false,
          "headline": "Revert \"L-ridge: restructure _linear_leaf_fit, bit-identical (GROW_PL…",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-18T16:40:34Z",
          "body_truncated": false,
          "is_coding_agent": false
        },
        {
          "oid": "332fd39839f02b9642a12bf6d4f5e0d657ff0e51",
          "body": "Row-major precomputed design table (kills the k double-indirections +\nisfinite branch per accumulate in the leaf-grouped loop), mirrored\nintercept column, hoisted h*x. Old kernel retained as\n_linear_leaf_fit_ref oracle; exact-equality test added; 455 tests incl.\nidentity goldens green. Timing read to follow.\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>",
          "is_bot": false,
          "headline": "L-ridge: restructure _linear_leaf_fit, bit-identical (GROW_PLAN Phase 1)",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-18T16:29:52Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "05c12237de7ec9d69958507734a0a069e56cc3a8",
          "body": "Binary ms/tree anomaly resolved = the linear-leaf ridge (6-24% of fit).\nL-ridge then L-pytree proceed; L-mc, L-leaf32, L-bin8 killed at measured\nceiling. Split kernel 37-58% of fit recorded as Phase-2 input.\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>",
          "is_bot": false,
          "headline": "GROW_PLAN Phase 0 complete: attribution table + lever order",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-18T16:24:55Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "fadde1260ae5bdeb4f6e9a558b57f8d60ca53da0",
          "body": "Kernel-level attribution via monkeypatched timers (split/descend/\nleaf_values/linear_fit/grow), ll-delta, thread-geometry, and dtype\nstream-width modes. No library change.\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>",
          "is_bot": false,
          "headline": "Phase-0 grow profiler: benchmarks/profile_grow.py (GROW_PLAN.md)",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-18T16:19:22Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "f16c7133a94ce16d167cd645c9b29db0b3592e17",
          "body": "Code survey done before registration (fused kernels, uint16 bins, oracle\ndiscipline all already in place). Phase 0 = in-tree attribution incl.\nthe binary 3.6-vs-1.5 ms/tree linear-leaf suspect; Phase 1 = bit-identical\nlevers only, ordered by measured ceilings; FP-drift kernels (histogram\nsubtraction) explicitly Phase-2 opt-in. Kill bar registered. PARETO_PLAN\nchecklist reconciled (step 1 closed via B-prep, M1 shipped in 0.17.0).\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>",
          "is_bot": false,
          "headline": "Step-3 grow-kernel program opened: pre-registered GROW_PLAN.md",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-18T15:55:01Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "aa9a2f5743720b38c6841c6544525bbb81dbc105",
          "body": "Elo 1256 -> 1267 (+11, flat inside the +-58 CI; pool sat still same-run\nso it is our-row noise). Train 0.84 -> 0.79 s/1K -- the multiclass\nselection tax never reaches Lite's median task. Frontier position\nunchanged: non-dominated between RandomForest and CatBoost.\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>",
          "is_bot": false,
          "headline": "tabarena_pareto.png refresh post-0.17.0 Lite re-read",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-18T12:44:03Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "66302ef612a7e86260ede99cec2046a4c16d60ee",
          "body": "Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>",
          "is_bot": false,
          "headline": "Release 0.17.0",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-18T01:17:46Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "92a13f329329aa1f89317125507dacede2f06c0c",
          "body": "Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>",
          "is_bot": false,
          "headline": "Merge m1-multiclass-cross: multiclass cross features SHIP",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-18T00:06:54Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "4d89abc1ec8f04ca61a25b65b6e4ad496e760b03",
          "body": "hc: canary 14/14, identity 13+13, okcupid Ens8 3/3 seeds positive both\nmetrics. gr: 59/59 exact ties. Gate: canary 36/36, eligible pooled F1\n+0.106% / Brier +1.360%, fit tax 1.92x summed. Verdict recorded in\nM1_PLAN.md.\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>",
          "is_bot": false,
          "headline": "M1 SHIP: tier-2 + gate PASS; docs, harness guard, CHANGELOG",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-18T00:06:45Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "f9fe3a6e45d8a7ce1021e3fb732225a1cc1b1e68",
          "body": "Identity perfect (125/125 untouched sets exact ties, both metrics).\nEligible slice at 3 seeds knife-edged the bar (7W-7L +0.61%, the 7th\nloss a -3e-5 pseudo-tie); pre-stated 6-seed extension resolves it:\nprimary 11W-4L +0.686%, Brier 9W-6L +0.719%, validity canaries exact.\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>",
          "is_bot": false,
          "headline": "M1 tier-1 screen: PASS (6-seed extension per registered deviation)",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-17T23:37:42Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "a24dc4fbc3170b878d304546d52263ad8f503381",
          "body": "_prep_matrices hoisted to _BaseBooster (encode-targets list), prep_cache\nplumbed into MulticlassBoosting.fit, classifier selection block unified\nover binary/multiclass via a _make factory. Same gates, budget, refit\nrule; NotImplementedError removed. 16 new tests; 454 green incl. goldens.\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>",
          "is_bot": false,
          "headline": "M1: multiclass cross features + raced selection (parity with binary)",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-17T23:28:21Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "37788e5de2953fe4196ac45b415d9865ef5f6507",
          "body": "Design locked before any library change: parity selection (same gates,\nCROSS_TOP_M, raced audition on softmax val logloss), prep_cache extension\nto MulticlassBoosting, auto-default parity. Treatment surface measured\nfirst: synth screen 15/34 multiclass sets eligible, hc 1/4 (okcupid-stem\nonly), OpenML gate 4/9, Grinsztajn zero (pure identity canary). Kill bars\nregistered.\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>",
          "is_bot": false,
          "headline": "M1 program opened: pre-registered plan for multiclass cross features",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-17T22:24:05Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "c347a032b9b6fa6ffeb609f6c7888e072c092af0",
          "body": "…ame-data bag; 93% pruning confirms collinearity but uniform averaging wins; B8 moot; bagging program fully closed\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>",
          "is_bot": false,
          "headline": "B7 VERDICT: KILL — probe 0/15 vs the shipped bag, -2.3% even vs the s…",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-17T20:24:17Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "de6dad5d477874e98298b39228963ea9d17ac0bb",
          "body": "Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>",
          "is_bot": false,
          "headline": "Release 0.16.1",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-17T20:13:10Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "aed205189ac8fa0ba168adb2b6f471e8a4be1eb9",
          "body": "…accuracy columns identical (Ens8 sweeps 100.0), absolute Chimera fit times down (gr -2.3%, hc -11.5%); in-chart slowdown 31.0x/6.4x vs 30.1x/6.3x is LightGBM-denominator box drift (its own raw time fell 6%), not a regression\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>",
          "is_bot": false,
          "headline": "pareto.png refresh post-B-prep (canonical 5-arm gr 20260717-153114): …",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-17T20:09:52Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "53df153a40609680503e8a246c5c8017efcb5228",
          "body": "Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>",
          "is_bot": false,
          "headline": "Merge bagging-bprep: B-prep SHIP (intra-fit prep reuse, bit-identical)",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-17T20:08:28Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "6a508e1d12a4cf4c056b10c91cc70b41707e08b5",
          "body": "… cheaper raw (single + every member); gr ~2%; checklist closed\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>",
          "is_bot": false,
          "headline": "B-prep VERDICT: SHIP — tier-2 identity 73/73 exact ties; hc fits ~12%…",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-17T20:08:21Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "0ac926b5ca09f19c3dd249d21c78f3dd53bb5e44",
          "body": "…ntra-fit prep reuse instead), smoke results, CHANGELOG\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>",
          "is_bot": false,
          "headline": "B-prep: design pivot recorded (cross-member border sharing SKIPPED; i…",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-17T19:32:21Z",
          "body_truncated": false,
          "is_coding_agent": true
        },
        {
          "oid": "42debf7d54ab425a5d43d75df0688ea49d0096d1",
          "body": "…are one preprocessing (bit-identical)\n\nThe selection auditions, the cross-augmented candidate, and the winner refit\nall recomputed identical prep (TS encodings, quantile borders, binning) 2-4x\nper fit. They now share a prep_cache keyed by cross_pairs: repeat fits reuse\nthe cached prep+matrices outr\n[…]\neen incl.\nall numerical-identity goldens; 7 new tests (prep-level splice identity,\nbooster cache-hit/augment identity, prep-runs-once counters).\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>",
          "is_bot": false,
          "headline": "B-prep: intra-fit prep reuse — booster fits within one sklearn fit sh…",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-17T19:28:36Z",
          "body_truncated": true,
          "is_coding_agent": true
        },
        {
          "oid": "b7028143057fa7e0058ebbb660d5d943108b1b11",
          "body": "….8 without replacement)\n\nCo-Authored-By: Claude Fable 5 <noreply@anthropic.com>",
          "is_bot": false,
          "headline": "docs: fix stale bootstrap-resamples wording (default is max_samples=0…",
          "author_name": "Nathan Walker",
          "author_login": "bbstats",
          "committed_at": "2026-07-17T19:12:47Z",
          "body_truncated": false,
          "is_coding_agent": true
        }
      ],
      "releases_count": 14,
      "commits_last_year": 367,
      "latest_release_at": "2026-07-24T16:21:51Z",
      "latest_release_tag": "v0.24.0",
      "releases_from_tags": false,
      "days_since_last_push": 0,
      "active_weeks_last_year": 8,
      "days_since_latest_release": 2,
      "mean_days_between_releases": 0.8
    },
    "community": {
      "has_readme": true,
      "has_license": true,
      "has_description": true,
      "has_contributing": false,
      "health_percentage": 57,
      "has_issue_template": false,
      "has_code_of_conduct": false,
      "has_pull_request_template": true
    },
    "ecosystem": {
      "packages": [
        {
          "name": "chimeraboost",
          "exists": true,
          "license": "Apache-2.0",
          "keywords": [
            "Development Status :: 4 - Beta",
            "Intended Audience :: Science/Research",
            "Programming Language :: Python :: 3",
            "Topic :: Scientific/Engineering :: Artificial Intelligence"
          ],
          "ecosystem": "pypi",
          "matches_repo": true,
          "registry_url": "https://pypi.org/project/chimeraboost/",
          "is_deprecated": false,
          "latest_version": "0.25.0",
          "repository_url": "https://github.com/bbstats/chimeraboost",
          "versions_count": 27,
          "total_downloads": null,
          "dependents_count": null,
          "deprecation_note": null,
          "maintainers_count": null,
          "monthly_downloads": 2036,
          "first_published_at": "2026-05-27T20:03:38.265872Z",
          "latest_published_at": "2026-07-26T16:40:55.207275Z",
          "latest_version_yanked": null,
          "days_since_latest_publish": 0
        }
      ]
    },
    "popularity": {
      "forks": 1,
      "stars": 7,
      "watchers": 0,
      "fork_history": {
        "days": [
          {
            "date": "2026-07-10",
            "count": 1
          }
        ],
        "complete": true,
        "collected": 1,
        "total_forks": 1
      },
      "star_history": null,
      "open_issues_and_prs": 1
    },
    "ai_readiness": {
      "has_nix": false,
      "example_dirs": [
        "examples"
      ],
      "has_llms_txt": false,
      "has_dockerfile": false,
      "has_mcp_signal": false,
      "bootstrap_files": [],
      "api_schema_files": [],
      "has_devcontainer": false,
      "typecheck_configs": [],
      "toolchain_manifests": [],
      "largest_source_bytes": 125169,
      "source_files_sampled": 125,
      "oversized_source_files": 3,
      "agent_instruction_files": [
        "CLAUDE.md"
      ],
      "agent_instruction_max_bytes": 5476
    },
    "dependencies": {
      "manifests": [
        "pyproject.toml",
        "requirements-docs.txt"
      ],
      "advisories": {
        "error": null,
        "scope": null,
        "source": null,
        "findings": [],
        "collected": false,
        "malicious": [],
        "truncated": false,
        "by_severity": {},
        "advisory_count": 0,
        "affected_count": 0,
        "assessed_count": 0,
        "malicious_count": 0,
        "assessed_package": null,
        "unassessed_count": 0,
        "direct_affected_count": 0
      },
      "ecosystems": [
        "pypi"
      ],
      "dependencies": [
        {
          "name": "numpy",
          "manifest": "pyproject.toml",
          "ecosystem": "pypi",
          "version_constraint": ">=1.22"
        },
        {
          "name": "numba",
          "manifest": "pyproject.toml",
          "ecosystem": "pypi",
          "version_constraint": ">=0.57"
        },
        {
          "name": "scikit-learn",
          "manifest": "pyproject.toml",
          "ecosystem": "pypi",
          "version_constraint": ">=1.0"
        },
        {
          "name": "scipy",
          "manifest": "pyproject.toml",
          "ecosystem": "pypi",
          "version_constraint": ">=1.7"
        }
      ],
      "all_dependencies": {
        "error": "GitHub dependency-graph SBOM unavailable (404); the dependency graph may be disabled for this repository",
        "source": null,
        "packages": [],
        "collected": false,
        "truncated": false,
        "total_count": null,
        "direct_count": null,
        "indirect_count": null
      }
    },
    "maintainership": {
      "issues": {
        "open_prs": 0,
        "merged_prs": 33,
        "open_issues": 1,
        "closed_ratio": 0,
        "closed_issues": 0,
        "closed_unmerged_prs": 0
      },
      "bus_factor": 1,
      "bot_contributors": 1,
      "top_contributors": [
        {
          "type": "User",
          "login": "bbstats",
          "commits": 364,
          "avatar_url": "https://avatars.githubusercontent.com/u/81766437?v=4"
        }
      ],
      "contributors_sampled": 1,
      "top_contributor_share": 1
    },
    "quality_signals": {
      "has_ci": true,
      "has_tests": true,
      "ci_workflows": [
        "docs.yml",
        "tests.yml"
      ],
      "has_docs_dir": true,
      "linter_configs": [],
      "has_editorconfig": false,
      "has_linter_config": false,
      "has_precommit_config": false
    },
    "security_signals": {
      "lockfiles": [],
      "scorecard": {
        "checks": [
          {
            "name": "Binary-Artifacts",
            "score": 10,
            "reason": "no binaries found in the repo",
            "documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#binary-artifacts"
          },
          {
            "name": "Branch-Protection",
            "score": null,
            "reason": "internal error: error during branchesHandler.setup: internal error: some github tokens can't read classic branch protection rules: https://github.com/ossf/scorecard-action/blob/main/docs/authentication/fine-grained-auth-token.md",
            "documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#branch-protection"
          },
          {
            "name": "CI-Tests",
            "score": 10,
            "reason": "6 out of 6 merged PRs checked by a CI test -- score normalized to 10",
            "documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#ci-tests"
          },
          {
            "name": "CII-Best-Practices",
            "score": 0,
            "reason": "no effort to earn an OpenSSF best practices badge detected",
            "documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#cii-best-practices"
          },
          {
            "name": "Code-Review",
            "score": 0,
            "reason": "Found 0/7 approved changesets -- score normalized to 0",
            "documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#code-review"
          },
          {
            "name": "Contributors",
            "score": 3,
            "reason": "project has 1 contributing companies or organizations -- score normalized to 3",
            "documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#contributors"
          },
          {
            "name": "Dangerous-Workflow",
            "score": 10,
            "reason": "no dangerous workflow patterns detected",
            "documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#dangerous-workflow"
          },
          {
            "name": "Dependency-Update-Tool",
            "score": 0,
            "reason": "no update tool detected",
            "documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#dependency-update-tool"
          },
          {
            "name": "Fuzzing",
            "score": 0,
            "reason": "project is not fuzzed",
            "documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#fuzzing"
          },
          {
            "name": "License",
            "score": 10,
            "reason": "license file detected",
            "documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#license"
          },
          {
            "name": "Maintained",
            "score": 0,
            "reason": "project was created within the last 90 days. Please review its contents carefully",
            "documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#maintained"
          },
          {
            "name": "Packaging",
            "score": null,
            "reason": "packaging workflow not detected",
            "documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#packaging"
          },
          {
            "name": "Pinned-Dependencies",
            "score": 0,
            "reason": "dependency not pinned by hash detected -- score normalized to 0",
            "documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#pinned-dependencies"
          },
          {
            "name": "SAST",
            "score": 0,
            "reason": "SAST tool is not run on all commits -- score normalized to 0",
            "documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#sast"
          },
          {
            "name": "Security-Policy",
            "score": 0,
            "reason": "security policy file not detected",
            "documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#security-policy"
          },
          {
            "name": "Signed-Releases",
            "score": null,
            "reason": "no releases found",
            "documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#signed-releases"
          },
          {
            "name": "Token-Permissions",
            "score": 0,
            "reason": "detected GitHub workflow tokens with excessive permissions",
            "documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#token-permissions"
          },
          {
            "name": "Vulnerabilities",
            "score": 10,
            "reason": "0 existing vulnerabilities detected",
            "documentation_url": "https://github.com/ossf/scorecard/blob/c395761df6afe1a69e476bc60a013a94bcbc153f/docs/checks.md#vulnerabilities"
          }
        ],
        "commit": "566573aa6c87bb8416be14b148699c26ef358423",
        "ran_at": "2026-07-26T16:41:46Z",
        "aggregate_score": 3.6,
        "scorecard_version": "v5.5.0"
      },
      "has_codeql_workflow": false,
      "has_security_policy": false,
      "has_dependabot_config": false
    },
    "contribution_flow": {
      "collected": true,
      "ci_last_run_at": "2026-07-26T16:37:48Z",
      "oldest_open_prs": [],
      "last_merged_pr_at": "2026-07-26T16:37:21Z",
      "ci_last_conclusion": "SUCCESS",
      "oldest_open_issues": [
        {
          "number": 34,
          "created_at": "2026-07-26T16:40:30Z",
          "last_comment_at": null,
          "last_comment_author": null
        }
      ]
    }
  },
  "config": {
    "disabled_metrics": [],
    "disabled_categories": [],
    "disabled_components": {}
  },
  "source": {
    "url": "https://github.com/bbstats/chimeraboost",
    "host": "github.com",
    "name": "chimeraboost",
    "owner": "bbstats"
  },
  "metrics": {
    "overall": {
      "key": "overall",
      "band": "moderate",
      "name": "Overall health",
      "note": null,
      "notes": [],
      "value": 54,
      "inputs": {
        "security": 36,
        "vitality": 76,
        "community": 39,
        "governance": 45,
        "engineering": 71
      },
      "components": []
    },
    "categories": [
      {
        "key": "vitality",
        "band": "good",
        "name": "Vitality",
        "value": 76,
        "weight": 0.22,
        "metrics": [
          {
            "key": "development_activity",
            "band": "moderate",
            "name": "Development activity",
            "note": null,
            "notes": [],
            "value": 60,
            "inputs": {
              "commits_last_year": 367,
              "human_commit_share": 0.99,
              "days_since_last_push": 0,
              "active_weeks_last_year": 8
            },
            "components": [
              {
                "key": "push_recency",
                "name": "Push recency",
                "detail": "last push 0 days ago",
                "points": 36,
                "status": "met",
                "details": [
                  {
                    "code": "push_recency",
                    "params": {
                      "days": 0
                    }
                  }
                ],
                "max_points": 36
              },
              {
                "key": "commit_cadence",
                "name": "Commit cadence",
                "detail": "8/52 weeks with commits",
                "points": 5.5,
                "status": "partial",
                "details": [
                  {
                    "code": "commit_cadence_weeks",
                    "params": {
                      "weeks": 8
                    }
                  }
                ],
                "max_points": 36
              },
              {
                "key": "commit_volume",
                "name": "Commit volume",
                "detail": "367 commits in the last year",
                "points": 18,
                "status": "met",
                "details": [
                  {
                    "code": "commits_last_year",
                    "params": {
                      "count": 367
                    }
                  }
                ],
                "max_points": 18
              },
              {
                "key": "openssf_scorecard_maintained",
                "name": "OpenSSF Scorecard: Maintained",
                "detail": "project was created within the last 90 days. Please review its contents carefully",
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 10
              }
            ]
          },
          {
            "key": "release_discipline",
            "band": "excellent",
            "name": "Release discipline",
            "note": "Excluded from scoring (no data or not applicable): OpenSSF Scorecard: Signed-Releases. Remaining weights renormalized.",
            "notes": [
              {
                "code": "excluded_no_data",
                "params": {
                  "components": [
                    "openssf_scorecard_signed_releases"
                  ]
                }
              },
              {
                "code": "weights_renormalized",
                "params": {}
              }
            ],
            "value": 100,
            "inputs": {
              "releases_count": 14,
              "latest_release_tag": "v0.24.0",
              "releases_from_tags": false,
              "days_since_latest_release": 2,
              "mean_days_between_releases": 0.8
            },
            "components": [
              {
                "key": "ships_releases",
                "name": "Ships releases",
                "detail": "14 releases published",
                "points": 27,
                "status": "met",
                "details": [
                  {
                    "code": "releases_published",
                    "params": {
                      "count": 14
                    }
                  }
                ],
                "max_points": 27
              },
              {
                "key": "release_recency",
                "name": "Release recency",
                "detail": "latest release 2 days ago",
                "points": 36,
                "status": "met",
                "details": [
                  {
                    "code": "release_recency",
                    "params": {
                      "days": 2
                    }
                  }
                ],
                "max_points": 36
              },
              {
                "key": "release_cadence",
                "name": "Release cadence",
                "detail": "a release every ~0.8 days",
                "points": 27,
                "status": "met",
                "details": [
                  {
                    "code": "release_cadence",
                    "params": {
                      "gap": 0.8
                    }
                  }
                ],
                "max_points": 27
              },
              {
                "key": "openssf_scorecard_signed_releases",
                "name": "OpenSSF Scorecard: Signed-Releases",
                "detail": "no releases found",
                "points": 0,
                "status": "excluded",
                "details": [
                  {
                    "code": "no_data",
                    "params": {}
                  }
                ],
                "max_points": 10
              }
            ]
          },
          {
            "key": "abandonment",
            "band": "excellent",
            "name": "Abandonment",
            "note": null,
            "notes": [],
            "value": 100,
            "inputs": {
              "cap": null,
              "state": "unverified",
              "guards": [],
              "signals": [],
              "red_flag": false,
              "multiplier_pct": 100,
              "declared_reason": null,
              "unverified_reason": "repository_too_young",
              "unanswered_open_prs": null,
              "unanswered_open_issues": null,
              "days_since_last_merged_pr": null,
              "days_since_last_human_commit": null,
              "days_since_last_human_commit_is_floor": false
            },
            "components": [
              {
                "key": "project_is_still_maintained",
                "name": "Project is still maintained",
                "detail": "maintenance record not established from the collected data",
                "points": 100,
                "status": "met",
                "details": [
                  {
                    "code": "abandonment_unverified",
                    "params": {}
                  }
                ],
                "max_points": 100
              }
            ]
          }
        ],
        "description": "Is the project alive — is code being written and are releases shipping?"
      },
      {
        "key": "community",
        "band": "at_risk",
        "name": "Community & Adoption",
        "value": 39,
        "weight": 0.18,
        "metrics": [
          {
            "key": "popularity",
            "band": "critical",
            "name": "Popularity & adoption",
            "note": null,
            "notes": [],
            "value": 13,
            "inputs": {
              "forks": 1,
              "stars": 7,
              "watchers": 0,
              "growth_state": "unverified",
              "growth_factor_pct": 100,
              "growth_unverified_reason": "no_history"
            },
            "components": [
              {
                "key": "stars",
                "name": "Stars",
                "detail": "7 stars",
                "points": 12.6,
                "status": "partial",
                "details": [
                  {
                    "code": "stars",
                    "params": {
                      "count": 7
                    }
                  }
                ],
                "max_points": 60
              },
              {
                "key": "forks",
                "name": "Forks",
                "detail": "1 forks",
                "points": 0,
                "status": "missed",
                "details": [
                  {
                    "code": "forks",
                    "params": {
                      "count": 1
                    }
                  }
                ],
                "max_points": 25
              },
              {
                "key": "watchers",
                "name": "Watchers",
                "detail": "0 watchers",
                "points": 0,
                "status": "missed",
                "details": [
                  {
                    "code": "watchers",
                    "params": {
                      "count": 0
                    }
                  }
                ],
                "max_points": 15
              }
            ]
          },
          {
            "key": "community_health",
            "band": "moderate",
            "name": "Community health",
            "note": null,
            "notes": [],
            "value": 57,
            "inputs": {
              "has_readme": true,
              "has_license": true,
              "has_contributing": false,
              "has_issue_template": false,
              "has_code_of_conduct": false,
              "has_pull_request_template": true
            },
            "components": [
              {
                "key": "readme",
                "name": "README",
                "detail": null,
                "points": 22.5,
                "status": "met",
                "details": [],
                "max_points": 22.5
              },
              {
                "key": "license",
                "name": "License",
                "detail": "recognized license (Apache-2.0)",
                "points": 22.5,
                "status": "met",
                "details": [
                  {
                    "code": "license_standard",
                    "params": {}
                  },
                  {
                    "code": "license_spdx",
                    "params": {
                      "spdx": "Apache-2.0"
                    }
                  }
                ],
                "max_points": 22.5
              },
              {
                "key": "contributing_guide",
                "name": "CONTRIBUTING guide",
                "detail": null,
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 18
              },
              {
                "key": "code_of_conduct",
                "name": "Code of conduct",
                "detail": null,
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 13.5
              },
              {
                "key": "issue_template",
                "name": "Issue template",
                "detail": null,
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 7.2
              },
              {
                "key": "pr_template",
                "name": "PR template",
                "detail": null,
                "points": 6.3,
                "status": "met",
                "details": [],
                "max_points": 6.3
              }
            ]
          },
          {
            "key": "ecosystem_adoption",
            "band": "moderate",
            "name": "Ecosystem adoption (downloads)",
            "note": "Excluded from scoring (no data or not applicable): Registry dependents. Remaining weights renormalized.",
            "notes": [
              {
                "code": "excluded_no_data",
                "params": {
                  "components": [
                    "registry_dependents"
                  ]
                }
              },
              {
                "code": "weights_renormalized",
                "params": {}
              }
            ],
            "value": 55,
            "inputs": {
              "packages": [
                "chimeraboost"
              ],
              "dependents": null,
              "ecosystems": "pypi",
              "total_downloads": null,
              "monthly_downloads": 2036
            },
            "components": [
              {
                "key": "monthly_downloads",
                "name": "Monthly downloads",
                "detail": "2,036 downloads/month across pypi",
                "points": 44.1,
                "status": "partial",
                "details": [
                  {
                    "code": "downloads_monthly",
                    "params": {
                      "count": 2036,
                      "ecosystems": "pypi"
                    }
                  }
                ],
                "max_points": 80
              },
              {
                "key": "registry_dependents",
                "name": "Registry dependents",
                "detail": "not reported by this ecosystem",
                "points": 0,
                "status": "excluded",
                "details": [
                  {
                    "code": "not_reported_by_this_ecosystem",
                    "params": {}
                  }
                ],
                "max_points": 20
              }
            ]
          }
        ],
        "description": "Does the project have users, downloads, attention, and a welcoming setup for contributors?"
      },
      {
        "key": "governance",
        "band": "at_risk",
        "name": "Sustainability & Governance",
        "value": 45,
        "weight": 0.24,
        "metrics": [
          {
            "key": "maintainer_resilience",
            "band": "critical",
            "name": "Maintainer resilience (bus factor)",
            "note": null,
            "notes": [],
            "value": 13,
            "inputs": {
              "bus_factor": 1,
              "contributors_sampled": 1,
              "top_contributor_share": 1
            },
            "components": [
              {
                "key": "bus_factor",
                "name": "Bus factor",
                "detail": "1 contributor(s) cover half of all commits",
                "points": 9,
                "status": "partial",
                "details": [
                  {
                    "code": "bus_factor",
                    "params": {
                      "count": 1
                    }
                  }
                ],
                "max_points": 54
              },
              {
                "key": "commit_distribution",
                "name": "Commit distribution",
                "detail": "top contributor authored 100% of commits",
                "points": 0,
                "status": "missed",
                "details": [
                  {
                    "code": "top_contributor_share",
                    "params": {
                      "share": 100
                    }
                  }
                ],
                "max_points": 22.5
              },
              {
                "key": "contributor_breadth",
                "name": "Contributor breadth",
                "detail": "1 contributors",
                "points": 1.4,
                "status": "partial",
                "details": [
                  {
                    "code": "contributors_sampled",
                    "params": {
                      "count": 1
                    }
                  }
                ],
                "max_points": 13.5
              },
              {
                "key": "openssf_scorecard_contributors",
                "name": "OpenSSF Scorecard: Contributors",
                "detail": "project has 1 contributing companies or organizations -- score normalized to 3",
                "points": 3,
                "status": "partial",
                "details": [],
                "max_points": 10
              }
            ]
          },
          {
            "key": "responsiveness",
            "band": "at_risk",
            "name": "Issue & PR responsiveness",
            "note": null,
            "notes": [],
            "value": 38,
            "inputs": {
              "merged_prs": 33,
              "open_issues": 1,
              "closed_issues": 0,
              "issue_closed_ratio": 0,
              "closed_unmerged_prs": 0
            },
            "components": [
              {
                "key": "issue_resolution",
                "name": "Issue resolution",
                "detail": "0% of issues closed",
                "points": 0,
                "status": "missed",
                "details": [
                  {
                    "code": "issues_closed_share",
                    "params": {
                      "share": 0
                    }
                  }
                ],
                "max_points": 46.75
              },
              {
                "key": "pr_acceptance",
                "name": "PR acceptance",
                "detail": "33/33 decided PRs merged",
                "points": 38.2,
                "status": "met",
                "details": [
                  {
                    "code": "decided_prs_merged",
                    "params": {
                      "merged": 33,
                      "decided": 33
                    }
                  }
                ],
                "max_points": 38.25
              },
              {
                "key": "openssf_scorecard_code_review",
                "name": "OpenSSF Scorecard: Code-Review",
                "detail": "Found 0/7 approved changesets -- score normalized to 0",
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 15
              }
            ]
          },
          {
            "key": "stewardship",
            "band": "at_risk",
            "name": "Ownership & stewardship",
            "note": "Excluded from scoring (no data or not applicable): Verified domain. Remaining weights renormalized.",
            "notes": [
              {
                "code": "excluded_no_data",
                "params": {
                  "components": [
                    "verified_domain"
                  ]
                }
              },
              {
                "code": "weights_renormalized",
                "params": {}
              }
            ],
            "value": 47,
            "inputs": {
              "followers": 13,
              "owner_type": "User",
              "is_verified": null,
              "owner_login": "bbstats",
              "public_repos": 13,
              "account_age_days": 1942
            },
            "components": [
              {
                "key": "ownership_backing",
                "name": "Ownership backing",
                "detail": "personal (user) account",
                "points": 10,
                "status": "partial",
                "details": [
                  {
                    "code": "owner_personal",
                    "params": {}
                  }
                ],
                "max_points": 30
              },
              {
                "key": "verified_domain",
                "name": "Verified domain",
                "detail": "not applicable to user accounts",
                "points": 0,
                "status": "excluded",
                "details": [
                  {
                    "code": "not_applicable_to_user_accounts",
                    "params": {}
                  }
                ],
                "max_points": 20
              },
              {
                "key": "owner_reach",
                "name": "Owner reach",
                "detail": "13 followers of bbstats",
                "points": 8.2,
                "status": "partial",
                "details": [
                  {
                    "code": "owner_followers",
                    "params": {
                      "count": 13,
                      "login": "bbstats"
                    }
                  }
                ],
                "max_points": 25
              },
              {
                "key": "track_record",
                "name": "Track record",
                "detail": "13 public repos, account ~5 yr old",
                "points": 19,
                "status": "partial",
                "details": [
                  {
                    "code": "public_repos",
                    "params": {
                      "count": 13
                    }
                  },
                  {
                    "code": "account_age_years",
                    "params": {
                      "years": 5
                    }
                  }
                ],
                "max_points": 25
              }
            ]
          },
          {
            "key": "package_maintenance",
            "band": "excellent",
            "name": "Package maintenance",
            "note": null,
            "notes": [],
            "value": 100,
            "inputs": {
              "packages": [
                "chimeraboost"
              ],
              "ecosystems": "pypi",
              "any_deprecated": false,
              "min_days_since_publish": 0
            },
            "components": [
              {
                "key": "published_resolvable",
                "name": "Published & resolvable",
                "detail": "1 package(s) on pypi",
                "points": 25,
                "status": "met",
                "details": [
                  {
                    "code": "packages_published",
                    "params": {
                      "count": 1,
                      "ecosystems": "pypi"
                    }
                  }
                ],
                "max_points": 25
              },
              {
                "key": "publish_recency",
                "name": "Publish recency",
                "detail": "latest publish 0 days ago",
                "points": 35,
                "status": "met",
                "details": [
                  {
                    "code": "publish_recency",
                    "params": {
                      "days": 0
                    }
                  }
                ],
                "max_points": 35
              },
              {
                "key": "version_history",
                "name": "Version history",
                "detail": "27 published versions",
                "points": 20,
                "status": "met",
                "details": [
                  {
                    "code": "published_versions",
                    "params": {
                      "count": 27
                    }
                  }
                ],
                "max_points": 20
              },
              {
                "key": "not_deprecated",
                "name": "Not deprecated",
                "detail": "active, not deprecated or yanked",
                "points": 20,
                "status": "met",
                "details": [
                  {
                    "code": "package_not_deprecated",
                    "params": {}
                  }
                ],
                "max_points": 20
              }
            ]
          }
        ],
        "description": "Will the project survive its people — bus factor, responsiveness, who backs it, and package upkeep?"
      },
      {
        "key": "engineering",
        "band": "good",
        "name": "Engineering Quality",
        "value": 71,
        "weight": 0.2,
        "metrics": [
          {
            "key": "engineering_practices",
            "band": "moderate",
            "name": "Engineering practices",
            "note": null,
            "notes": [],
            "value": 68,
            "inputs": {
              "has_ci": true,
              "has_tests": true,
              "has_editorconfig": false,
              "has_linter_config": false,
              "has_precommit_config": false
            },
            "components": [
              {
                "key": "ci_workflows",
                "name": "CI workflows",
                "detail": "2 workflow(s)",
                "points": 24,
                "status": "met",
                "details": [
                  {
                    "code": "ci_workflows",
                    "params": {
                      "count": 2
                    }
                  }
                ],
                "max_points": 24
              },
              {
                "key": "tests_present",
                "name": "Tests present",
                "detail": null,
                "points": 24,
                "status": "met",
                "details": [],
                "max_points": 24
              },
              {
                "key": "linter_config",
                "name": "Linter config",
                "detail": null,
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 16
              },
              {
                "key": "pre_commit_hooks",
                "name": "Pre-commit hooks",
                "detail": null,
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 9.6
              },
              {
                "key": "editorconfig",
                "name": ".editorconfig",
                "detail": null,
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 6.4
              },
              {
                "key": "openssf_scorecard_ci_tests",
                "name": "OpenSSF Scorecard: CI-Tests",
                "detail": "6 out of 6 merged PRs checked by a CI test -- score normalized to 10",
                "points": 20,
                "status": "met",
                "details": [],
                "max_points": 20
              }
            ]
          },
          {
            "key": "documentation",
            "band": "good",
            "name": "Documentation",
            "note": null,
            "notes": [],
            "value": 75,
            "inputs": {
              "topics": [],
              "has_wiki": true,
              "homepage": null,
              "has_readme": true,
              "has_docs_dir": true,
              "has_description": true
            },
            "components": [
              {
                "key": "readme",
                "name": "README",
                "detail": null,
                "points": 30,
                "status": "met",
                "details": [],
                "max_points": 30
              },
              {
                "key": "documentation_directory",
                "name": "Documentation directory",
                "detail": null,
                "points": 25,
                "status": "met",
                "details": [],
                "max_points": 25
              },
              {
                "key": "documentation_homepage_site",
                "name": "Documentation / homepage site",
                "detail": null,
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 15
              },
              {
                "key": "repository_description",
                "name": "Repository description",
                "detail": null,
                "points": 10,
                "status": "met",
                "details": [],
                "max_points": 10
              },
              {
                "key": "topics",
                "name": "Topics",
                "detail": null,
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 10
              },
              {
                "key": "wiki",
                "name": "Wiki",
                "detail": null,
                "points": 10,
                "status": "met",
                "details": [],
                "max_points": 10
              }
            ]
          }
        ],
        "description": "Are baseline engineering and documentation practices in place?"
      },
      {
        "key": "security",
        "band": "at_risk",
        "name": "Security",
        "value": 36,
        "weight": 0.16,
        "metrics": [
          {
            "key": "security_posture",
            "band": "at_risk",
            "name": "Security posture",
            "note": "Excluded from scoring (no data or not applicable): Branch-Protection, Packaging, Signed-Releases. Remaining weights renormalized.",
            "notes": [
              {
                "code": "excluded_no_data",
                "params": {
                  "components": [
                    "branch_protection",
                    "packaging",
                    "signed_releases"
                  ]
                }
              },
              {
                "code": "weights_renormalized",
                "params": {}
              }
            ],
            "value": 36,
            "inputs": {
              "source": "openssf_scorecard",
              "checks_evaluated": 15,
              "scorecard_version": "v5.5.0",
              "checks_inconclusive": 3,
              "scorecard_aggregate": 3.6
            },
            "components": [
              {
                "key": "binary_artifacts",
                "name": "Binary-Artifacts",
                "detail": "no binaries found in the repo",
                "points": 7.5,
                "status": "met",
                "details": [],
                "max_points": 7.5
              },
              {
                "key": "branch_protection",
                "name": "Branch-Protection",
                "detail": "internal error: error during branchesHandler.setup: internal error: some github tokens can't read classic branch protection rules: https://github.com/ossf/scorecard-action/blob/main/docs/authentication/fine-grained-auth-token.md",
                "points": 0,
                "status": "excluded",
                "details": [
                  {
                    "code": "no_data",
                    "params": {}
                  }
                ],
                "max_points": 7.5
              },
              {
                "key": "ci_tests",
                "name": "CI-Tests",
                "detail": "6 out of 6 merged PRs checked by a CI test -- score normalized to 10",
                "points": 2.5,
                "status": "met",
                "details": [],
                "max_points": 2.5
              },
              {
                "key": "cii_best_practices",
                "name": "CII-Best-Practices",
                "detail": "no effort to earn an OpenSSF best practices badge detected",
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 2.5
              },
              {
                "key": "code_review",
                "name": "Code-Review",
                "detail": "Found 0/7 approved changesets -- score normalized to 0",
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 7.5
              },
              {
                "key": "contributors",
                "name": "Contributors",
                "detail": "project has 1 contributing companies or organizations -- score normalized to 3",
                "points": 0.8,
                "status": "partial",
                "details": [],
                "max_points": 2.5
              },
              {
                "key": "dangerous_workflow",
                "name": "Dangerous-Workflow",
                "detail": "no dangerous workflow patterns detected",
                "points": 10,
                "status": "met",
                "details": [],
                "max_points": 10
              },
              {
                "key": "dependency_update_tool",
                "name": "Dependency-Update-Tool",
                "detail": "no update tool detected",
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 7.5
              },
              {
                "key": "fuzzing",
                "name": "Fuzzing",
                "detail": "project is not fuzzed",
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 5
              },
              {
                "key": "license",
                "name": "License",
                "detail": "license file detected",
                "points": 2.5,
                "status": "met",
                "details": [],
                "max_points": 2.5
              },
              {
                "key": "maintained",
                "name": "Maintained",
                "detail": "project was created within the last 90 days. Please review its contents carefully",
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 7.5
              },
              {
                "key": "packaging",
                "name": "Packaging",
                "detail": "packaging workflow not detected",
                "points": 0,
                "status": "excluded",
                "details": [
                  {
                    "code": "no_data",
                    "params": {}
                  }
                ],
                "max_points": 5
              },
              {
                "key": "pinned_dependencies",
                "name": "Pinned-Dependencies",
                "detail": "dependency not pinned by hash detected -- score normalized to 0",
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 5
              },
              {
                "key": "sast",
                "name": "SAST",
                "detail": "SAST tool is not run on all commits -- score normalized to 0",
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 5
              },
              {
                "key": "security_policy",
                "name": "Security-Policy",
                "detail": "security policy file not detected",
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 5
              },
              {
                "key": "signed_releases",
                "name": "Signed-Releases",
                "detail": "no releases found",
                "points": 0,
                "status": "excluded",
                "details": [
                  {
                    "code": "no_data",
                    "params": {}
                  }
                ],
                "max_points": 7.5
              },
              {
                "key": "token_permissions",
                "name": "Token-Permissions",
                "detail": "detected GitHub workflow tokens with excessive permissions",
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 7.5
              },
              {
                "key": "vulnerabilities",
                "name": "Vulnerabilities",
                "detail": "0 existing vulnerabilities detected",
                "points": 7.5,
                "status": "met",
                "details": [],
                "max_points": 7.5
              }
            ]
          }
        ],
        "description": "Are visible security and supply-chain practices strong, with no malicious dependency and no unresolved high-risk jurisdiction exposure?"
      },
      {
        "key": "ai_readiness",
        "band": "moderate",
        "name": "AI Readiness",
        "value": 52,
        "weight": 0,
        "metrics": [
          {
            "key": "ai_agent_context",
            "band": "excellent",
            "name": "Agent context & guidance",
            "note": null,
            "notes": [],
            "value": 85,
            "inputs": {
              "has_llms_txt": false,
              "legible_history_share": 0.939,
              "agent_instruction_files": [
                "CLAUDE.md"
              ],
              "agent_instruction_max_bytes": 5476
            },
            "components": [
              {
                "key": "agent_instructions",
                "name": "Agent instructions",
                "detail": "CLAUDE.md",
                "points": 45,
                "status": "met",
                "details": [
                  {
                    "code": "file_list",
                    "params": {
                      "files": "CLAUDE.md"
                    }
                  }
                ],
                "max_points": 45
              },
              {
                "key": "machine_readable_docs_llms_txt",
                "name": "Machine-readable docs (llms.txt)",
                "detail": null,
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 15
              },
              {
                "key": "legible_commit_history",
                "name": "Legible commit history",
                "detail": "93 of 99 human commits state their intent (structured subject or explanatory body)",
                "points": 40,
                "status": "met",
                "details": [
                  {
                    "code": "legible_history",
                    "params": {
                      "legible": 93,
                      "sampled": 99
                    }
                  }
                ],
                "max_points": 40
              }
            ]
          },
          {
            "key": "ai_verify_loop",
            "band": "at_risk",
            "name": "Verify loop (build / test / typecheck)",
            "note": null,
            "notes": [],
            "value": 32,
            "inputs": {
              "has_nix": false,
              "has_tests": true,
              "lockfiles": [],
              "has_dockerfile": false,
              "typed_language": false,
              "bootstrap_files": [],
              "has_devcontainer": false,
              "has_linter_config": false,
              "typecheck_configs": [],
              "agent_commit_share": 0.85,
              "toolchain_manifests": [],
              "dependency_bot_commit_share": 0
            },
            "components": [
              {
                "key": "one_command_bootstrap",
                "name": "One-command bootstrap",
                "detail": null,
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 18
              },
              {
                "key": "automated_tests",
                "name": "Automated tests",
                "detail": null,
                "points": 22,
                "status": "met",
                "details": [],
                "max_points": 22
              },
              {
                "key": "lint_format_config",
                "name": "Lint / format config",
                "detail": null,
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 11
              },
              {
                "key": "static_type_checking",
                "name": "Static type checking",
                "detail": null,
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 11
              },
              {
                "key": "reproducible_environment",
                "name": "Reproducible environment",
                "detail": null,
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 10
              },
              {
                "key": "demonstrated_agent_practice",
                "name": "Demonstrated agent practice",
                "detail": "85 of the last 100 commits agent-authored or agent-credited",
                "points": 10,
                "status": "met",
                "details": [
                  {
                    "code": "agent_authored_commits",
                    "params": {
                      "count": 85,
                      "sampled": 100
                    }
                  }
                ],
                "max_points": 10
              },
              {
                "key": "automated_maintenance",
                "name": "Automated maintenance",
                "detail": "no automated dependency updates observed",
                "points": 0,
                "status": "missed",
                "details": [
                  {
                    "code": "no_dependency_automation",
                    "params": {}
                  }
                ],
                "max_points": 8
              },
              {
                "key": "openssf_scorecard_pinned_dependencies",
                "name": "OpenSSF Scorecard: Pinned-Dependencies",
                "detail": "dependency not pinned by hash detected -- score normalized to 0",
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 10
              }
            ]
          },
          {
            "key": "ai_code_legibility",
            "band": "moderate",
            "name": "Code legibility for models",
            "note": null,
            "notes": [],
            "value": 54,
            "inputs": {
              "primary_language": "Python",
              "largest_source_bytes": 125169,
              "source_files_sampled": 125,
              "oversized_source_files": 3
            },
            "components": [
              {
                "key": "type_checkable_code",
                "name": "Type-checkable code",
                "detail": "Python without a type-check config",
                "points": 0,
                "status": "missed",
                "details": [
                  {
                    "code": "no_typecheck_config_language",
                    "params": {
                      "language": "Python"
                    }
                  }
                ],
                "max_points": 45
              },
              {
                "key": "manageable_file_sizes",
                "name": "Manageable file sizes",
                "detail": "3/125 source files over 60KB",
                "points": 53.7,
                "status": "partial",
                "details": [
                  {
                    "code": "oversized_source_files",
                    "params": {
                      "kb": 60,
                      "sampled": 125,
                      "oversized": 3
                    }
                  }
                ],
                "max_points": 55
              }
            ]
          },
          {
            "key": "ai_interfaces",
            "band": "at_risk",
            "name": "Machine-readable interfaces",
            "note": null,
            "notes": [],
            "value": 40,
            "inputs": {
              "example_dirs": [
                "examples"
              ],
              "has_mcp_signal": false,
              "api_schema_files": []
            },
            "components": [
              {
                "key": "api_schema_openapi_graphql_proto",
                "name": "API schema (OpenAPI/GraphQL/proto)",
                "detail": null,
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 40
              },
              {
                "key": "mcp_server",
                "name": "MCP server",
                "detail": null,
                "points": 0,
                "status": "missed",
                "details": [],
                "max_points": 20
              },
              {
                "key": "runnable_examples",
                "name": "Runnable examples",
                "detail": "examples",
                "points": 40,
                "status": "met",
                "details": [
                  {
                    "code": "file_list",
                    "params": {
                      "files": "examples"
                    }
                  }
                ],
                "max_points": 40
              }
            ]
          }
        ],
        "description": "How well is the repo equipped to be developed and maintained with AI coding agents? An independent, experimental badge — weight 0.0, so it is surfaced on its own and does not affect the overall health score."
      }
    ],
    "metrics_version": "1.13.0"
  },
  "warnings": [
    "Star history unavailable: GitHub GraphQL error: Resource not accessible by personal access token",
    "GitHub dependency-graph SBOM unavailable (404); the dependency graph may be disabled for this repository",
    "deps.dev does not index pypi:chimeraboost@0.25.0; advisories assessed against the repository dependency graph instead"
  ],
  "report_type": "repository",
  "generated_at": "2026-07-26T16:41:59.574188Z",
  "schema_version": "0.27.0",
  "badge_url": "https://raw.githubusercontent.com/inspect-software/badges/main/v1/b/bbstats/chimeraboost.svg",
  "full_name": "bbstats/chimeraboost",
  "license_state": "standard",
  "license_spdx": "Apache-2.0"
}