Commit Graph
75 Commits
Author SHA1 Message Date
maximeandClaude Opus 5.5 82edc98a29 auto: NORDEND-169 names NORDEND-310 as a Perturbation (its new family); FLORIDA-18 no longer points to a return label that does not exist (found by the validation pass)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-18 10:38:36 +01:00
maximeandClaude Opus 5.5 9ccd04d9a5 auto: 36 Displacement labels that came back requalified Perturbation (review kept, a change of family the user asked for); 6 closing sentences reworded to their family, review removed (NEBRASKA-3, NORDEND-257, NORDEND-310, FLORIDA-18, WILDSTRUBEL-136, NORDEND-250); with the reviews aimsight-app had written in these files since the last commit
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-18 10:20:13 +01:00
maximeandClaude Opus 5.5 df3a7120d3 Families: Displacement is something that moved and never came back; Perturbation, new, what comes back without another family's cause, nothing to do (spec 2026-11-18-displacement-split-design)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-18 10:19:02 +01:00
maximeandClaude Opus 5.5 c89c1b9976 Rules: a missing point is not labeled for itself (the station filtered it), No data only when the station delivers nothing, thermal movement only when the corrected curves keep it (replaces D-117), Target issue only for a target rejected almost always; the mm curves follow the same rule and threshold
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-18 07:16:59 +01:00
maximeandClaude Opus 5.5 764af7d8ec auto: the long labels requalified as analyses like the others, <STATION>_<start>_<end>/ with request.json (ex long-labels/<STATION>/); their labels carry the analysis; work/ out of git
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-18 07:16:59 +01:00
maximeandClaude Opus 5.5 ddb17c2189 auto: labels requalified under the 5-noise rules (only what stands out from the noise): 32 analyses refined, the 6 paused periods verified, the long labels requalified with the fixes the period analysts found; 1153 -> 961 labels
- 32 analyses: each label kept, rewritten or removed by its analyst (section Requalification of the report; mm magnitudes where the targets have them)
- SNCF01 x3, WILDSTRUBEL 2026-09-01 -> 2026-10-06 x3: verified from their existing labels under the new rules
- long labels (no analysis): auto/analyses/long-labels/<STATION>/; NORDEND-169 and -195 run to 07/11, so NORDEND-283 and -284 (their continuation) are removed
- reading-rules.md, stations.md: the citations of removed or rewritten labels point to their successor, or to [expert]

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-17 23:58:00 +01:00
maximeandClaude Opus 5.5 c779de7720 Rules: a movement of the corr is labeled only beyond 5 own noises, the threshold of the thermal cycle; a missing point at any amplitude; under it, the ordinary noise goes in the analysis, not in a label (WYOMING: an agitation under the threshold, removed according to the expert)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-17 15:30:03 +01:00
maximeandClaude Opus 5.5 2a5d87659b auto: the analysts verified the labels of 32 periods (11 stations, up to 14 days each): 710 labels -> 835, 587 unchanged; one analysis folder per period with its report (section Verification)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-17 15:13:44 +01:00
maximeandClaude Opus 5.5 df979301b5 auto/support/: the folders of the support agent, work/ outside git
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-17 14:52:08 +01:00
maximeandClaude Opus 5.5 528336f9e5 auto: the WYOMING analysis of 6-16 November and its labels; three old analyses removed (LENZSPITZE 1 Nov, NORDEND 4-7 and 5 Nov)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-17 12:25:48 +01:00
maximeandClaude Opus 5.5 4fd623b3e1 auto: review set to 2026-11-17 on every label
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-17 10:31:38 +01:00
maximeandClaude Opus 5.5 bed0f7d09e review is now the date the user validated a label in aimsight-app; the agents' review dates removed from auto/
Quality over quantity: only the analyses discussed and validated in the app are marked. validate.py no longer
requires review; precedents.py gives the provenance as validated by the user or not.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-17 10:01:22 +01:00
maximeandClaude Opus 5.5 542e91ccb2 For aimsight-app: validate.check (the rules of one label), client.py (what the client is told of an event, from families.md); families.md: For the client per variety
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-17 09:32:20 +01:00
maximeandClaude Opus 5.5 83f92bb7b9 families: Obstruction comes from visibility.py (brief) and target_issue.py (target obscured)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-16 17:43:17 +01:00
maximeandClaude Opus 5.5 69cba19cdc Dépôts renommés avec le préfixe aimsight- : chemins et noms mis à jour
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-16 14:56:41 +01:00
maximeandClaude Opus 5.5 7591488173 NORDEND-243: the error_value of CTR02 and PR01 rises with the jolt at 12:00 (checked on the data), not 'nothing more'
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-16 13:26:57 +01:00
maximeandClaude Opus 5.5 f34616ae38 The labeling scripts in scripts/ (validate, ids, precedents); the index stays in .index/
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-16 13:20:31 +01:00
maximeandClaude Opus 5.5 4688d8ab9e Analyst knowledge: reading-rules.md, stations.md, a phenomenon catalogue in families.md; the label map and knowledge/ archived
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-16 12:47:01 +01:00
maximeandClaude Opus 5.5 1c8479d962 Analyst knowledge: design and plan (one source per kind of knowledge)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-16 12:40:54 +01:00
maximeandClaude Opus 5.5 68a0eefd8b English: label map, families, labeling rules, knowledge/, validate.py (ex controle.py), ids.py, precedents.py; the labels of auto/ translated, keys start/end/family/review/analysis (manual/: keys only); analysis folders renamed; the 2026-11-16 trial analysis (NORDEND-243)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-16 12:40:03 +01:00
maximeandClaude Opus 5.5 4d926ecddc precedents.py démarre le conteneur Ollama s'il ne répond pas, puis relance ; testé
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-16 12:13:17 +01:00
maximeandClaude Opus 5.5 ac86d8f6cc Le labeler archivé (conception et plan, jamais construit) : hors de l'architecture du 2026-11-16
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-16 11:46:22 +01:00
maximeandClaude Opus 5.5 41071fa486 Notes : les outils de lecture dans tools/scripts/
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-16 11:45:31 +01:00
maximeandClaude Opus 5.5 6dcc1fbc8e Les précédents ne lisent que auto/ : manual/ est à l'utilisateur seul, aucun processus ne s'en sert
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-16 11:37:50 +01:00
maximeandClaude Opus 5.5 f379836cdc Notes : les outils de lecture sont dans le dépôt tools
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-16 10:51:34 +01:00
maximeandClaude Opus 5.5 1d9bfc20b3 precedents.py : le RAG des étiquettes (bge-m3 par Ollama), index .index/ hors de git
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-16 09:55:48 +01:00
maximeandClaude Opus 5.5 1f32cf8198 Plan de construction du labeler (5 tâches), validé, à exécuter
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-14 15:44:01 +01:00
maximeandClaude Opus 5.5 65acde61c2 Conception du labeler (étiquettes à faible effort, versées dans auto/ avec source labeler) ; CLAUDE.md et permissions du dépôt
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-14 15:39:02 +01:00
maximeandClaude Opus 5.5 6067f8dfcd Cohérence des dépôts : controle.py et ids.py se lancent d'ici ou par le lien ; notes sans renvois morts (pompage dans aimsight-data, règles non négociables recopiées, analyste de Diagnostic-App)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-13 17:01:14 +01:00
maximeandClaude Opus 5.5 b465d5c99b Dépôt des étiquettes, repris de Support-Agent et aligné sur Diagnostic-App (462ba9e)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-13 16:25:54 +01:00
maximeandClaude Opus 5.5 fc5bdafd73 App : l'historique des analyses dans la barre latérale ; une analyse interrompue se reprend
Analyse NORDEND, 04.11 21:12 - 07.11 14:12 : toile d'araignée sur CTR03, 14 étiquettes non relues.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-12 17:34:15 +01:00
maximeandClaude Opus 5.5 9c56e3c0f1 App Data : Analysis remplace Labelling, l'étiqueteur par claude -p sur une plage, avec dialogue (D-147)
Premier essai : LENZSPITZE, 01.11 10:30-11:40 (train de 11:00), 4 étiquettes non relues.
CLAUDE.md : LibreDesk traité dans un autre projet.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-12 17:00:51 +01:00
maximeandClaude Opus 5.5 620f536e9e Label ids first in each JSON line (D-131)
ids.py puts the id first and never re-gives an id retired in the same operation; labeling.py
keeps it first on add and edit; controle.py requires it first. Cross-review apply: a modified
label keeps its id, and revue is today's date. All 991 auto labels reordered, content unchanged.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-05 08:58:47 +01:00
maximeandClaude Opus 5.5 d9833b3fdf Auto labels take what the expert's labels know where they conflicted, compiled against the data (D-130)
JUNGFRAU (the PS_DES_COUDRES night, 07.10), NORDEND (03.09 web over CTR04), LENZSPITZE
(turbulence beyond the bridge), WILDSTRUBEL (well compensated, small residual at the bottom).
Expert labels unchanged; map section 6 updated.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-05 08:56:28 +01:00
maximeandClaude Opus 5.5 d4520f9f46 Stable ids for auto labels, a summary for the label map, reading tools improved (D-131)
- Every auto label carries an id (<STATION>-<n>, labeling/ids.py), kept on edit, given to
  new labels by labeling.py and by the reviews; controle.py requires it, unique. The
  cross-review matches labels by id, and an applied review is locked (applique.md).
- The label map starts with a summary (labels-map/sommaire.py): readers take the summary and
  sections 2-4 in full, then only the cards of the phenomena they meet.
- Tool-builder session on the reviewers' difficulties: jours.py --lissage/--fraction,
  ensemble.py --suivre and rejected measures left out of --rigide, --residus non, zero noise
  after a lost reference, new agent-tools/stations.py (several stations on one timeline),
  long outputs convention in lire-les-donnees.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-05 08:49:54 +01:00
maximeandClaude Opus 5.5 7c0cdee8cf Labeling processes: lessons from the auto-label reviews (D-130)
- The labeler writes to its own folder (etiquettes.jsonl); labeling/auto/ only holds
  reviewed labels, merged by /review-auto-labels (new apply script: dated, sorted, checked).
- labeling/controle.py also checks a proposed file; the labeler and reviewers run it before
  handing back. Reviewers keep their measurements in revue.md, never in the labels.
- Reviewer instructions explicitly add their scripts, .venv/bin/python and for loops to the
  lire-les-donnees commands (the contradiction raised by 12 reviews).
- Expert validation optional in /review-auto-labels; an expert hint given to the labeler is
  written in the labels it touches. Cross-review ends with the releveur.
- docs/ameliorations-processus.md: 17 entries from this review, by the releveur.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-05 08:31:53 +01:00
maximeandClaude Opus 5.5 14ad6ea3f9 Auto labels reviewed per station and cross-reviewed; all labeling knowledge in labeling/ (D-130)
- Every station's auto labels reviewed by an analyst (/review-auto-labels), then a
  cross-review per family across all stations (new skill /cross-review-labels),
  a finishing pass, and descriptions rewritten in words (no numeric values).
- The three never-labeled ranges (WILDSTRUBEL 07/09-01/10, NORDEND 16/08-01/10)
  labeled and reviewed: 991 auto labels on 11 stations, labeling/controle.py at 0 errors.
- Labeling knowledge gathered in labeling/: regles-d-etiquetage.md (new), familles.md
  (settled boundaries), connaissances/ (notes moved from docs/, links updated),
  the label map rebuilt from reviewed labels, events and cases; events tied to their
  stations (PS_DES_COUDRES is JUNGFRAU). The labeler and reviewers read only labeling/.
- labels-map reads auto labels and events; decision D-130; living plan points 73-75.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-05 08:21:57 +01:00
maximeandClaude Opus 5.5 4ae7afd89d Image quality metrics in labeling: qualite.py, LENZSPITZE October labels reviewed with them (D-129)
Definitions of peak, flux, sharpness, fwhm, dominance and snr in docs/domaine;
agent-tools/qualite.py (ratio to a calm reference, per label, outside labels,
figures), inventaire/graphique/resume aware of the six columns, lire-les-donnees
and the labeler's instructions updated. LENZSPITZE 16.10-01.11 relabeled: 67
labels, 4 found by the metrics.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-04 13:34:04 +01:00
maximeandClaude Opus 5.5 6202b5e2e9 Labeling: the six quality metrics in the plot menu, after Camera movement, as in Explore
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-04 12:33:36 +01:00
maximeandClaude Opus 5.5 af273e1708 Coherence of BREITHORN, MICHIGAN and NEBRASKA October labels (same building, per the expert)
One analyst checks neighbouring stations together (.claude/skills/review-auto-labels/coherence.md). The month's drift, shared by the three cameras (same step on 08/10, same recoil 13-16/10), is Displacement everywhere, cause not established (the building or its temperature); the 09/10 reference losses and the 15-16/10 gap are tied across stations, a common cause to look for in the building; the 26/10 cold morning added on BREITHORN and NEBRASKA; MICHIGAN 27/10 as Displacement. Proposed per-station files versioned.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-04 09:28:19 +01:00
maximeandClaude Opus 5.5 bacc06e37c Labeler: lessons from the three October reviews (BREITHORN, MICHIGAN, NEBRASKA)
Bounds checked target by target on both columns; a slowing drift stays one event; a phenomenon found once is searched over the whole batch; "the camera, not the targets" checked with ensemble.py --rigide; housing temperature, "d'après l'expert" in words. familles.md: a small error_value rise at the housing temperature trough, nothing on the images, is Thermal displacement, not Visibility. Map lessons left to the next /labels-map.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-04 09:06:44 +01:00
maximeandClaude Opus 5.5 c5a49ec269 MICHIGAN and NEBRASKA October labels reviewed by the analyst (/review-auto-labels), all suggestions applied
MICHIGAN: 13 labels, 3 kept, 10 corrected (the 05/10 reconfiguration as Target issue, one drift label from 03/10 counting the compensated thermal movement, the housing temperature instead of the factory's). NEBRASKA: 8 labels, 2 kept, 6 corrected, 2 missed added (COLJ19 alone on 17/10, the cold morning of 19/10); cold mornings as Thermal displacement instead of Visibility. All carry "revue": "2026-11-04"; lessons for the labeler not applied yet.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-04 09:04:18 +01:00
maximeandClaude Opus 5.5 2d3e6dffad Reviewed labels: "revue" is the review date, as for manual labels, and is not reviewed again
All 26 BREITHORN October labels carry "revue": "2026-11-04"; the expert's thermal answer is said in the description. /review-auto-labels skips labels that already carry "revue" (count, analyst review, replacement); /labels-map reads all reviewed auto labels (D-128).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-04 08:46:06 +01:00
maximeandClaude Opus 5.5 8192ef580f Review of the labeler's labels: /review-auto-labels (D-128); BREITHORN October reviewed
The analyst reviews labeling/auto/ labels on a range chosen by the user (bounds, one event, family, description, missed ones) and proposes corrected labels; the expert answers its questions, then the range is replaced. First run on BREITHORN October: 10 corrected, 25 and 26 merged, one missed night added; the expert confirms the camera's cold-night deviations are thermal (housing temperature, not air). Labels the expert settled carry "revue", read by /labels-map.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-04 08:40:27 +01:00
maximeandClaude Opus 5.5 13cf28431f Etiqueteur: October labels for 8 stations; its folder moves to labeling/auto/etiqueteur/, analyses and scripts versioned (D-126)
Labels for CLARIDEN, FLORIDA, JUNGFRAU (gaps 06-20.10, 26.10-01.11), LENZSPITZE,
NIEDERHORN, NORDEND, SNCF01, WILDSTRUBEL. Figures and tool outputs stay out of git.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-03 16:33:41 +01:00
maximeandClaude Sonnet 5.5 57ab5b3a09 Labeling: no hover on bars, lighter zone background
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
2026-11-03 15:32:29 +01:00
maximeandClaude Opus 5.5 8754d439c0 JUNGFRAU labels get their family; lone jumps scored as Displacement (D-120)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-03 15:16:21 +01:00
maximeandClaude Opus 5.5 4b2fcfd213 One detector per family, family keys equal to their names (D-119)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-03 14:57:52 +01:00
maximeandClaude Opus 5.5 39429837db Labels: the trou family and corr_x/corr_y as the anomaly (D-111); thermal movement is always labeled, the camera's included even when compensated (D-117)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-03 14:13:15 +01:00
maximeandClaude Opus 5.5 3bee717014 Detector-learner (D-116): the detectors learn from the labeler's labels alone, without events; critic, upgrader and control, the control guarded by the score on the other labeled stations (regression.py); NEBRASKA and NORDEND labels
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-03 13:46:36 +01:00