- 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>
35 lines
2.1 KiB
Python
35 lines
2.1 KiB
Python
"""Profil jour moyen de la pose de la caméra, par heure UTC, sur les jours d'étiquettes thermiques d'une station.
|
|
Pour chaque jour civil UTC et chaque axe (pan, tilt, rotation, scale) : la médiane à l'heure, ramenée de 0 (plus bas du jour)
|
|
à 100 (plus haut du jour) ; puis la médiane sur les jours, heure par heure. Une suite d'heures au même niveau : la caméra
|
|
s'y repose ; une pente régulière jusqu'à l'aube : elle revient toute la nuit. Amplitude du jour en bruits horaires.
|
|
Usage, depuis la racine : python3 <ce fichier> STATION [AXE] (n'écrit rien)"""
|
|
import csv, glob, json, sys, statistics as st
|
|
from datetime import datetime, timedelta
|
|
AX = ['pan', 'tilt', 'rotation', 'scale']
|
|
def t(s): return datetime.fromisoformat(s.replace('Z', '+00:00').replace(' ', 'T'))
|
|
station = sys.argv[1]
|
|
h = {}
|
|
for p in sorted(glob.glob(f'data/{station}/camera/*.csv')):
|
|
for r in csv.DictReader(open(p)):
|
|
try: v = [float(r['median_' + a]) for a in AX]
|
|
except (ValueError, KeyError): continue
|
|
h.setdefault(t(r['time_utc']).replace(minute=0, second=0), []).append(v)
|
|
P = {k: [st.median(x[i] for x in l) for i in range(4)] for k, l in h.items()}
|
|
ks = sorted(P)
|
|
bruit = [st.median(abs(P[b][i] - P[a][i]) for a, b in zip(ks, ks[1:]) if b - a == timedelta(hours=1)) or 1e-9 for i in range(4)]
|
|
jours = sorted({t(json.loads(l)['debut']).date() for l in open(f'labeling/auto/{station}.jsonl')
|
|
if l.strip() and json.loads(l)['categorie'] == 'Thermal displacement'})
|
|
axes = [AX.index(sys.argv[2])] if len(sys.argv) > 2 else range(4)
|
|
for i in axes:
|
|
prof = {hh: [] for hh in range(24)}; amps = []
|
|
for j in jours:
|
|
d = {k.hour: P[k][i] for k in ks if k.date() == j}
|
|
if len(d) < 20: continue
|
|
lo, hi = min(d.values()), max(d.values())
|
|
if hi == lo: continue
|
|
amps.append((hi - lo) / bruit[i])
|
|
for hh, v in d.items(): prof[hh].append(100 * (v - lo) / (hi - lo))
|
|
if not amps: continue
|
|
print(f'{station} {AX[i]:8s} ampl. {st.median(amps):4.0f} bruits, {len(amps)} j :',
|
|
' '.join(f'{hh:02d}:{st.median(prof[hh]):3.0f}' if prof[hh] else f'{hh:02d}: -' for hh in range(24)))
|