Skip to content

Fix(#14122): purge CS1701/CS0649 — tranche famille A Probas/Infer (4 carnets) #21727

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
myia-ai-01 merged 6 commits into from
Oct 21, 2026
Merged

Fix(#14122): purge CS1701/CS0649 — tranche famille A Probas/Infer (4 carnets) #21727

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
3d26bc5
Fix(#14122): purge warnings CS1701/CS0649 — tranche famille A Probas/…
jsboige Oct 21, 2026
d655c4b
Attestation twin -- 3 paires Probas/Infer (#14122, PR #21727)
jsboige Oct 21, 2026
ab3b4e5
Merge remote-tracking branch 'origin/main' into fix/14122-infer-cs1701
jsboige Oct 21, 2026
fed502f
Merge origin/main into fix/14122-infer-cs1701
jsboige Oct 21, 2026
bdac302
Merge branch 'main' into fix/14122-infer-cs1701
jsboige Oct 21, 2026
931f8a0
Fix(twin,#14122): renuméroter les attestations 0020 -> 0021 (collisio…
jsboige Oct 21, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
527 changes: 266 additions & 261 deletions MyIA.AI.Notebooks/Probas/Infer/Infer-01-Setup.ipynb
View file
Open in desktop

Large diffs are not rendered by default.

805 changes: 456 additions & 349 deletions MyIA.AI.Notebooks/Probas/Infer/Infer-01b-Premiers-Modeles.ipynb
View file
Open in desktop

Large diffs are not rendered by default.

1,802 changes: 1,297 additions & 505 deletions MyIA.AI.Notebooks/Probas/Infer/Infer-02b-Debugging-Bonnes-Pratiques.ipynb
View file
Open in desktop

Large diffs are not rendered by default.

1,492 changes: 1,364 additions & 128 deletions MyIA.AI.Notebooks/Probas/Infer/Infer-03-Factor-Graphs.ipynb
View file
Open in desktop

Large diffs are not rendered by default.

1 change: 1 addition & 0 deletions scripts/notebook_tools/twin_pairs.d/probas-1-setup.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,6 +6,7 @@
bridge_verdict: SOTA-OK
bridge_verdict_reason: "Modele nominal lib-vs-lib cite par owner #10382 (\"PyMC vs Infer, 2 moteurs qui brillent\") : C# = Microsoft Infer.NET (Microsoft.ML.Probabilistic, NuGet officiel, native .NET, inférence variationnelle EP/VMP) et Python = PyMC (pip pymc, native Python, NUTS/HMC + ADVI). Chaque cote atteint le moteur SOTA de son ecosysteme probabiliste -- ni pont intermediaire ni reimplementation. Aucune asymetrie de machinerie a corriger. Verifie firsthand : C# Microsoft.ML.Probabilistic refs + execution_count, Python pymc imports."
known_differences:
- '2026-10-21 (PR #21727, myia-po-2028:CoursIA) : REBASELINE cote C# -- tranche famille A #14122 (avertissement CS1701). Fix causal : `display(HTML(FactorGraphHelper.GetLatestFactorGraphHtml()))` devient `FactorGraphHelper.GetLatestFactorGraphHtml().DisplayAs("text/html")` -- meme HTML rendu sans lier `IHtmlContent` (Microsoft.AspNetCore.Html.Abstractions 2.2 vs hote 8.0) ; 2 sites de code (cellules 23 et 31) + 3 mentions didactiques alignees (docstring cellule 35, WriteLine cellule 10, markdown cellule 36 -- les eleves ne doivent pas recopier la forme qui declenche le warning). CS1701 en sortie : 1 -> 0. Re-execution complete du carnet (12/12 cellules de code, 0 erreur, 0 execution_count null, kernel .net-csharp, metadata.papermill 27,72 s) ; SVG factor-graph conserve (aucun repli viz-js, #15209 evite). Cote Python byte-identique a main. Attestation = nouveau blob SHA + content_csharp_sha du cote C#.'
- '2026-10-07 (myia-po-2027:CoursIA) #19778 : csharp_sha rebaseline -- table Navigation cellule 0,
colonne Suivant : Infer-2-Gaussian-Mixtures -> Infer-1b-Premiers-Modeles (raccord de la chaine
1 -> 1b -> 2, entree orpheline 1b). Infer-1b n a pas de jumeau Python enregistre -- chaine nav
Expand Down
6 changes: 6 additions & 0 deletions scripts/notebook_tools/twin_pairs.d/probas-1-setup/0019-2026-10-21-myia-po-2028-CoursIA.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,6 @@
date: '2026-10-21'
by: myia-po-2028:CoursIA
python_sha: 13932cfc8b5dd990e4bd96c950154e3c2502edb7
csharp_sha: 3551d741ddc567a35b2d79f486293aec95360452
content_python_sha: e3485fa8888c9dd51d23bdb9f208a32882b5a24cb8beb16547c50239108b8393
content_csharp_sha: a91eb6867786fdb02abc275c50b4946a5cbe6a580e149e49f4b4b1b3710bec39
1 change: 1 addition & 0 deletions scripts/notebook_tools/twin_pairs.d/probas-2b-debugging.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,6 +6,7 @@
bridge_verdict: SOTA-OK
bridge_verdict_reason: "Modele nominal lib-vs-lib cite par owner #10382 (\"PyMC vs Infer, 2 moteurs qui brillent\") : C# = Microsoft Infer.NET (Microsoft.ML.Probabilistic, NuGet officiel, native .NET, inférence variationnelle EP/VMP) et Python = PyMC (pip pymc, native Python, NUTS/HMC + ADVI). Chaque cote atteint le moteur SOTA de son ecosysteme probabiliste -- ni pont intermediaire ni reimplementation. Aucune asymetrie de machinerie a corriger. Verifie firsthand : C# Microsoft.ML.Probabilistic refs + execution_count, Python pymc imports."
known_differences:
- '2026-10-21 (PR #21727, myia-po-2028:CoursIA) : REBASELINE cote C# -- tranche famille A #14122. Meme fix causal que Probas-1 Setup : `.DisplayAs("text/html")` remplace `display(HTML(...))` sur 3 sites de code (cellules 10, 28, 41) + 2 mentions didactiques (cellules 6 et 23). CS1701 en sortie : 3 -> 0. Re-execution complete (19/19 cellules de code, 0 erreur, 0 execution_count null, kernel .net-csharp, metadata.papermill 41,76 s) ; les sorties NuGet vides de la base deviennent des `Installed Packages` reelles. Cote Python byte-identique a main. Attestation = nouveau blob SHA + content_csharp_sha du cote C#.'
- '2026-10-02 (myia-po-2023:CoursIA) #18793 : REBASELINE cote C# -- cure canonique d''accents #2876 markdown-only (Infer-2b-Debugging-Bonnes-Pratiques.ipynb : 5 cellules md, 0 cellule code, outputs/execution_count intacts) ; jumeau Python non touche. Aucun strip outille en attente.'
- '2026-08-21 (rebaseline asymetrique myia-po-2024:CoursIA-2) : PR #12065 corrige heading_in_list cote C# (Infer-6-Debugging.ipynb, 3 findings) tandis que le jumeau Python (PyMC-06-Debugging.ipynb) etait deja a 0 finding. La convergence unilaterale est legitime (cf ai-01 arbitrage 2026-08-21 15:00Z, DM msg-20260821T130006-lc9u2c) : la parite reelle s''ameliore, elle n''est pas enshrinee. Aucun strip outille en attente (probe banner 0, papermill 0, machine paths 0).'
- 'Socle pedagogique commun : debugging de modeles probabilistes (diagnostics d''inference) modelise dans les deux stacks.'
Expand Down
6 changes: 6 additions & 0 deletions ...notebook_tools/twin_pairs.d/probas-2b-debugging/0021-2026-10-21-myia-po-2028-CoursIA.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,6 @@
date: '2026-10-21'
by: myia-po-2028:CoursIA
python_sha: 926b9d141cca9e01d0636ac59a2e315a365c11a8
csharp_sha: 9ad797a5b531e0b34f8b8dece5bfa86a258ad853
content_python_sha: e1108f0be702cafbf59c33b6e1ab17831446b2073e6f4b752023cc68e89eb78d
content_csharp_sha: f114d4f1914114d8955e5143056a3c032c8e0c01a44591f42c823ef3a06aecf4
1 change: 1 addition & 0 deletions scripts/notebook_tools/twin_pairs.d/probas-3-factor-graphs.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,6 +6,7 @@
bridge_verdict: SOTA-OK
bridge_verdict_reason: "Modele nominal lib-vs-lib cite par owner #10382 (\"PyMC vs Infer, 2 moteurs qui brillent\") : C# = Microsoft Infer.NET (Microsoft.ML.Probabilistic, NuGet officiel, native .NET, inférence variationnelle EP/VMP) et Python = PyMC (pip pymc, native Python, NUTS/HMC + ADVI). Chaque cote atteint le moteur SOTA de son ecosysteme probabiliste -- ni pont intermediaire ni reimplementation. Aucune asymetrie de machinerie a corriger. Verifie firsthand : C# Microsoft.ML.Probabilistic refs + execution_count, Python pymc imports."
known_differences:
- '2026-10-21 (PR #21727, myia-po-2028:CoursIA) : REBASELINE cote C# -- tranche famille A #14122. Meme fix causal : `.DisplayAs("text/html")` sur 2 sites de code (cellules 18 et 30) + 1 mention didactique (cellule 5). CS1701 en sortie : 2 -> 0. Re-execution complete (15/15 cellules de code, 0 erreur, 0 execution_count null, kernel .net-csharp, metadata.papermill 29,13 s) ; SVG factor-graphs conserves. Cote Python byte-identique a main. Attestation = nouveau blob SHA + content_csharp_sha du cote C#.'
- '2026-10-18 (myia-po-2024:CoursIA-2) #20246 : REBASELINE (c.494) -- dedup 6 doublons d''attestations twin (c.493) avait deplace le git blob SHA des 2 jumeaux par retrait des attestations obsoletes de la lane (collision index NNNN avec sessions cloud/po-2027). --update pour re-attester la parite (la cause de la collision -- plusieurs sessions concurrentes sur le meme index du journal -- est aujourd''hui resolue, le registre ne porte plus qu''une attestation par paire par date). Cellules code byte-identiques au commit precedent ; aucun strip outille ulterieur prevu (cf #8957).'
- '2026-09-27 (myia-po-2024:CoursIA-2) #18028 : REBASELINE cote Python -- chains=4 rendu explicite sur les appels pm.sample, valeur deja effective (les sorties portent "4 chains in 4 jobs"), resultats inchanges ; jumeau C# non touche. Sorties par ailleurs restaurees a celles de main : la re-execution de cette branche tournait sans compilateur g++ et laissait fuiter l avertissement machine dans les sorties.'
- '2026-09-23 (myia-po-2026:CoursIA) #17304 : REBASELINE cote Python apres reparation audit Hermes #17222 (3 findings : API pt.switch, numerotation exercices 1-2-3, jargon CPT defini) -- markdown-only, cellules code byte-identiques, outputs et execution_count preserves (exception C.2). content_csharp_sha inchange : le jumeau C# n''est pas touche.'
Expand Down
6 changes: 6 additions & 0 deletions ...ebook_tools/twin_pairs.d/probas-3-factor-graphs/0021-2026-10-21-myia-po-2028-CoursIA.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,6 @@
date: '2026-10-21'
by: myia-po-2028:CoursIA
python_sha: 192eb086d55ef5c50cba8e289ecc5c8d04d29f20
csharp_sha: 3dbe4b776d62e300c4b650de64b34a7823c6aaed
content_python_sha: 3d9a99fdb033f4dfc850480af7854992b72526c3e6c24142bed81df915d014d1
content_csharp_sha: 65adb0728c017247b26456129510960e2b0282ede75b84390a491abccf19fb5f
Loading