Files
Support-Agent/labeling/auto/revue-croisee/2026-11-04/visibility/outils/alternance.py
T
maximeandClaude Opus 5.5 35b5ed4d74 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

33 lines
1.8 KiB
Python

"""Le bruit de jour est-il un bruit rapide (turbulence) ou une courbe lente mal compensée ?
Pour chaque cible, les pas d'une mesure à la suivante (Δcorr, mesures acceptées, cadence régulière) ;
corrélation de deux pas qui se suivent : vers -0,5 pour un bruit blanc (le point revient), vers 0 ou plus
pour une dérive ou une vague. Médiane sur les cibles, par tranche d'heures UTC données.
Usage (depuis la racine) : python3 alternance.py STATION DEBUT FIN H1-H2 [H1-H2 ...]
"""
import sys, statistics as st
sys.path.insert(0, "agent-tools")
import donnees as d
nom, t1, t2 = sys.argv[1], d.instant(sys.argv[2]), d.instant(sys.argv[3])
tranches = [tuple(map(int, a.split("-"))) for a in sys.argv[4:]]
dans = lambda h, a, b: a <= h < b if a < b else (h >= a or h < b)
par = d.charger(f"data/{nom}")
cib = {s: [m for m in d.filtrer(ms, t1, t2) if m["corr_x"] is not None] for s, ms in par.items() if not d.est_reference(ms)}
ts = sorted({m["t"] for ms in cib.values() for m in ms})
cad = st.median([(b - a).total_seconds() for a, b in zip(ts, ts[1:])])
for a, b in tranches:
rs, ampl = [], []
for s, ms in cib.items():
for q in ("corr_x", "corr_y"):
p1, p2 = [], []
for m0, m1, m2 in zip(ms, ms[1:], ms[2:]):
if (m2["t"] - m0["t"]).total_seconds() > 2.5 * cad or not dans(m1["t"].hour, a, b):
continue
p1.append(m1[q] - m0[q]); p2.append(m2[q] - m1[q])
if len(p1) > 20 and st.pstdev(p1) > 0 and st.pstdev(p2) > 0:
rs.append(st.correlation(p1, p2)); ampl.append(st.median(abs(x) for x in p1))
print(f"{nom} {a:02d}-{b:02d} UTC : corrélation de deux pas voisins {st.median(rs):+.2f} "
f"(médiane de {len(rs)} séries), pas médian {st.median(ampl):.4f}")