File size: 1,335 Bytes
ac5db64 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 | """Simulate final worry levels from modal_artifacts/ml_probe.json raw dumps
(dev tooling — runs the same triage path as the pipeline, no GPU needed)."""
from __future__ import annotations
import json
import os
import sys
sys.path.insert(0, os.path.join(os.path.dirname(os.path.dirname(os.path.abspath(__file__))), "src"))
if hasattr(sys.stdout, "reconfigure"):
sys.stdout.reconfigure(encoding="utf-8")
from guardian import triage # noqa: E402
def main() -> None:
with open("modal_artifacts/ml_probe.json", encoding="utf-8") as f:
data = json.load(f)
for p in data:
parsed = triage.parse_model_json(p["raw"])
if parsed is None:
print(f"{p['name']} ({p['degradation']}): PARSE FAIL")
continue
ex = triage.validate_extraction(parsed)
text = " | ".join(filter(None, [
ex.sender, ex.what_they_want, ex.requested_action, ex.what_this_is,
*ex.key_facts, *ex.worry_reasons,
*[s.evidence for s in ex.scam_signals]]))
signals = triage.merge_signals(ex.scam_signals, triage.run_heuristics(text))
level = triage.worry_level(ex.document_type, signals)
print(f"{p['name']} ({p['degradation']}): level={level} "
f"signals={[s.id for s in signals]}")
if __name__ == "__main__":
main()
|