wf-improve scripts take their paths as arguments (--labels, --data)

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
maximeandClaude Opus 5.5 committed 2026-11-13 12:38:29 +01:00
1 parent 80e00a4898
commit cff2b0ad1e
3 files changed
+40 -42

No files matched your search

+17 -17
View File
@@ -1,15 +1,14 @@
"""Benchmark of the wf-improve workflow (D-116): the detectors against all the labeller's labels (labeling/auto/). """Benchmark of the detectors against a reference truth: every labelled station.
For each station of labeling/auto/ that has its folder in data/, the score per family and per instant of score.py For each <STATION>.jsonl of --labels that has its folder <STATION>/ in --data, the score per family and per instant of
(D-112), over the range of its labels: from the day of the first to the midnight after the last. Then the total, family score.py, over the range of its labels: from the day of the first to the midnight after the last. Then the total,
by family, which says whether the detectors do better overall from one session to the next. The block of the session's family by family, which says whether the detectors do better overall. The block of a station is its score; the
station is its score; the other blocks, the non-regression: a family that goes back on a station is a regression, even others, the non-regression: a family that goes back on a station is a regression, even if the total rises.
if the total rises. Usage: python3 regression.py --labels <folder of <STATION>.jsonl> --data <folder of <STATION>/> [--stations A,B]
Usage: python3 detectors/skills/wf-improve/scripts/regression.py [--stations A,B]
""" """
import argparse import argparse
import json
from datetime import timedelta from datetime import timedelta
from pathlib import Path
import score # next to it import score # next to it
@@ -17,31 +16,32 @@ from data import iso, parse_time # on the path through score
def label_range(path): def label_range(path):
"""labeling/auto/<STATION>.jsonl -> (start, end) iso, in whole UTC days, end excluded.""" """Labels JSONL -> (start, end) iso, in whole UTC days, end excluded."""
with open(path, encoding="utf-8") as f: labels = score.read_labels(path)
labels = [json.loads(l) for l in f if l.strip()]
start = min(e["debut"] for e in labels)[:10] start = min(e["debut"] for e in labels)[:10]
# ponytail: the end of the batch is not written; midnight closest to the last bound (a label bound overflows by one # ponytail: the end of the labelled batch is not written; midnight closest to the last bound (a label bound
# measurement after midnight); the real end of the batch in labeling/auto/ if the gap matters # overflows by one measurement after midnight)
end = iso(parse_time(max(e["fin"] for e in labels)) + timedelta(hours=12))[:10] end = iso(parse_time(max(e["fin"] for e in labels)) + timedelta(hours=12))[:10]
return iso(parse_time(start)), iso(parse_time(end)) return iso(parse_time(start)), iso(parse_time(end))
if __name__ == "__main__": if __name__ == "__main__":
p = argparse.ArgumentParser(usage=__doc__) p = argparse.ArgumentParser(usage=__doc__)
p.add_argument("--labels", required=True, help="folder of the labels, one <STATION>.jsonl per station")
p.add_argument("--data", required=True, help="folder of the station folders, <STATION>/")
p.add_argument("--stations", help="A,B: these stations only (a shorter run)") p.add_argument("--stations", help="A,B: these stations only (a shorter run)")
a = p.parse_args() a = p.parse_args()
total = {} total = {}
for path in sorted((score.WORK / "labeling" / "auto").glob("*.jsonl")): for path in sorted(Path(a.labels).glob("*.jsonl")):
station = path.stem station = path.stem
if a.stations and station not in a.stations.split(","): if a.stations and station not in a.stations.split(","):
continue continue
folder = score.WORK / "data" / station folder = Path(a.data) / station
if not folder.is_dir(): if not folder.is_dir():
print(f"{station}: no data/{station}/, skipped\n") print(f"{station}: no {folder}/, skipped\n")
continue continue
start, end = label_range(path) start, end = label_range(path)
n, labels = score.family_scores(str(folder), start, end) n, labels = score.family_scores(str(folder), str(path), start, end)
score.show(f"{station}: {start} → {end}, {len(labels)} labels", n) score.show(f"{station}: {start} → {end}, {len(labels)} labels", n)
print() print()
for k, v in n.items(): for k, v in n.items():
+22 -24
View File
@@ -1,14 +1,14 @@
"""Score of the detectors on a range: the detectors' events against the labels of labeling/auto/<STATION>.jsonl that """Score of the detectors on a range: the detectors' events against the labels touching the range, per family and per
touch the range, per family (labeling/familles.md) and per instant (D-112). An instant: an hour of the range where at instant. An instant: a time of the range where at least one sensor has a measurement; it receives, on one side, the
least one sensor has a measurement; it receives, on one side, the families of the labels covering it, bounds excluded families of the labels covering it, bounds excluded (except for a hole, without a measurement inside), on the other,
(except for a hole, without a measurement inside), on the other, those of the detectors' events (categories.py, the those of the detectors' events (categories.py, the range classified with its 3 days of history). The category of an
range classified with its 3 days of history). The category of an event is its family. event is its family.
found: labelled instants that the detectors also flagged (share of the labelled ones); found: labelled instants that the detectors also flagged (share of the labelled ones);
correct: flagged instants that a label also puts in the family (share of the flagged ones). correct: flagged instants that a label also puts in the family (share of the flagged ones).
The score is a fact, not a verdict: the labels can be wrong too. No label on the range: exits with an error. The score is a fact, not a verdict: the labels can be wrong too. No label on the range: exits with an error.
Label fields (written by Diagnostic-app): debut, fin, categorie (the family), description. Labels: a JSONL file, one label per line, {"id", "debut", "fin", "categorie", "description"} (UTC ISO times, categorie =
Run from the working repository, where detectors/ is mounted. the family); its bounds are the healthy measurement before and after the anomaly.
Usage: python3 detectors/skills/wf-improve/scripts/score.py <station folder> --from YYYY-MM-DD --to YYYY-MM-DD Usage: python3 score.py <station folder> --labels <labels.jsonl> --from YYYY-MM-DD --to YYYY-MM-DD
""" """
import argparse import argparse
import json import json
@@ -17,13 +17,11 @@ import sys
from datetime import timedelta from datetime import timedelta
from pathlib import Path from pathlib import Path
REPO = Path(os.path.abspath(__file__)).parents[3] # the detectors repository sys.path.insert(0, str(Path(__file__).resolve().parents[3] / "scripts")) # the detectors
WORK = REPO.parent # the working repository (labeling/, data/)
sys.path.insert(0, str(REPO / "scripts"))
import categories # noqa: E402 import categories # noqa: E402
from data import iso, load, parse_time # noqa: E402 from data import iso, load, parse_time # noqa: E402
HISTORY = timedelta(days=3) # as the batch command (D-056) HISTORY = timedelta(days=3) # as the batch command of categories.py
def events(measurements, start, end, folder=None): def events(measurements, start, end, folder=None):
@@ -43,25 +41,24 @@ def count(pairs):
return n return n
def auto_labels(station, start, end): def read_labels(path, start=None, end=None):
"""labeling/auto/<STATION>.jsonl -> the labeller's labels touching [start, end).""" """Labels JSONL -> the labels touching [start, end) (all without bounds); [] if the file does not exist."""
path = WORK / "labeling" / "auto" / f"{station}.jsonl" if not os.path.exists(path):
if not path.exists():
return [] return []
with open(path, encoding="utf-8") as f: with open(path, encoding="utf-8") as f:
return [e for e in map(json.loads, filter(str.strip, f)) if e["debut"] < end and e["fin"] >= start] return [e for e in map(json.loads, filter(str.strip, f))
if (end is None or e["debut"] < end) and (start is None or e["fin"] >= start)]
def family_scores(folder, start, end): def family_scores(folder, labels_path, start, end):
"""Score of a range per family and per instant -> ({family: [labelled, flagged, both]}, labels).""" """Score of a range per family and per instant -> ({family: [labelled, flagged, both]}, labels)."""
# ponytail: per instant and not per sensor, the labels only give their sensors in the text; a sensors field in the # ponytail: per instant and not per sensor, the labels only give their sensors in the text; a sensors field in the
# label if this precision is missing # label if this precision is missing
measurements = load(folder) measurements = load(folder)
labels = auto_labels(os.path.basename(os.path.normpath(folder)), start, end) labels = read_labels(labels_path, start, end)
evts = events(measurements, start, end, folder) evts = events(measurements, start, end, folder)
instants = sorted({iso(m["t"]) for ms in measurements.values() for m in ms if start <= iso(m["t"]) < end}) instants = sorted({iso(m["t"]) for ms in measurements.values() for m in ms if start <= iso(m["t"]) < end})
# bounds excluded: they are the healthy measurement before and after (D-111); a hole, without a measurement # bounds excluded: they are the healthy measurement before and after; a hole, without a measurement inside, keeps them
# inside, keeps them
empty = {i for i, e in enumerate(labels) if not any(e["debut"] < t < e["fin"] for t in instants)} empty = {i for i, e in enumerate(labels) if not any(e["debut"] < t < e["fin"] for t in instants)}
pairs = [] pairs = []
for t in instants: for t in instants:
@@ -87,11 +84,12 @@ def show(title, n):
if __name__ == "__main__": if __name__ == "__main__":
p = argparse.ArgumentParser(usage=__doc__) p = argparse.ArgumentParser(usage=__doc__)
p.add_argument("folder") p.add_argument("folder")
p.add_argument("--labels", required=True)
p.add_argument("--from", dest="start", required=True) p.add_argument("--from", dest="start", required=True)
p.add_argument("--to", dest="end", required=True) p.add_argument("--to", dest="end", required=True)
a = p.parse_args() a = p.parse_args()
station, start, end = os.path.basename(os.path.normpath(a.folder)), iso(parse_time(a.start)), iso(parse_time(a.end)) station, start, end = os.path.basename(os.path.normpath(a.folder)), iso(parse_time(a.start)), iso(parse_time(a.end))
n, labels = family_scores(a.folder, start, end) n, labels = family_scores(a.folder, a.labels, start, end)
if not labels: if not labels:
sys.exit(f"No label of {station} on {start} → {end} in labeling/auto/{station}.jsonl: analyse the range first.") sys.exit(f"No label of {station} on {start} → {end} in {a.labels}.")
show(f"{station}: {start} → {end}, {len(labels)} labels of labeling/auto/{station}.jsonl, per family and per instant", n) show(f"{station}: {start} → {end}, {len(labels)} labels of {a.labels}, per family and per instant", n)
+1 -1
View File
@@ -1,7 +1,7 @@
"""For the upgrader: the cadence of the measurements of the range, the number of targets and references, then each event """For the upgrader: the cadence of the measurements of the range, the number of targets and references, then each event
of the range with its sensors (categories.py only gives their number). Same computation as categories.py without of the range with its sensors (categories.py only gives their number). Same computation as categories.py without
--chunk (3 days of history). --chunk (3 days of history).
Usage: python3 detectors/skills/wf-improve/scripts/sensors.py <folder> --from YYYY-MM-DD --to YYYY-MM-DD Usage: python3 sensors.py <folder> --from YYYY-MM-DD --to YYYY-MM-DD
""" """
import argparse import argparse
import statistics import statistics