Files
aimsight-labeling/auto/analyses/long-labels/CLARIDEN/analysis.md
T
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

1.6 KiB
Raw Blame History

Requalification of the long labels: CLARIDEN

  • CLARIDEN-1 (Target issue, P-300-B, 01/10 → 01/11): kept as is. P-300-B is rejected at 4070 of the 4070 measurement instants present in the range (inventory.py --start 2026-10-01T00:00Z --end 2026-11-01T00:00Z; 6371 of 10137 over the whole period, since 29/09). A target rejected almost always: Target issue.
  • CLARIDEN-14 (Displacement, 06/10 22:00 → 11/10 06:00): corrected (description). The drift is beyond the threshold: ranges.py --k 5 from 05/10 10:30 to 15/10 22:50, the level taken outside the label: D-301-C corr_x out of level up to −19.5 noises (−0.22 px, about −1.1 mm) until 08/10 09:20 before it rises to its later level; D-303-B corr_y up to −15.6 noises (about −0.44 mm) until 08/10 08:40; D-304-C corr_x −9.0, D-303-C corr_y −7.1, D-306-B corr_x +6.4 noises. Bounds and family hold. Correction applied (found by the analyst of the 01/10–11/10 period): "No camera movement" is wrong, the camera tilt drifts over these days and D-303-B, D-303-C, D-304-C, D-301-C follow it (hourly r −0.86 to −0.95, CLARIDEN_2026-10-01T0000_2026-10-11T0000/analysis.md); the conclusion becomes "camera to monitor". Review removed.

Sources read: tools/GUIDE.md, tools/DATA.md, labeling/families.md, labeling/labeling-rules.md, labeling/reading-rules.md, labeling/stations.md, input.jsonl, the analyses of the three CLARIDEN periods (lines on CLARIDEN-1 and CLARIDEN-14). Commands: inventory.py (month, work/inv_month.txt), ranges.py (corr and corr_mm, 5 targets, --outside the label).