romfast/rar-autopass
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

refactor: comentarii strict functionale, fara referinte PRD/stories

Browse Source 
Curatare globala a comentariilor si docstring-urilor (app, tools, teste,
scripturi): eliminate referintele la PRD-uri, US-xxx, task-uri istorice si
review-uri; pastrata doar informatia functionala, formulata scurt. Regula
adaugata in CLAUDE.md (sectiunea Stil). Fara modificari de cod sau comportament.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
This commit is contained in:
Claude Agent
2026-07-06 13:48:38 +00:00
parent 86408887e5
commit 44f261e269
226 changed files with 1313 additions and 1760 deletions
Download Patch File Download Diff File Expand all files Collapse all files

20
tools/mapare-llm/or_label.py
View File

@@ -1,19 +1,19 @@
"""Etichetator batch offline OpenRouter (Layer 1) — L14-S1.
"""Etichetator batch offline OpenRouter (Layer 1).
Clasifica denumirile de operatii service in cele 18 coduri RAR + NUL.
Cerinte implementate (PRD 5.14 / Decision Audit Trail):
Cerinte implementate:
1. Prioritizare pe FRECVENTA (desc): corpus_by_freq() din or_common
2. Grupare pe similaritate (rapidfuzz token_sort_ratio, threshold conservator
Eng-F7): LLM eticheteaza doar reprezentantul, codul se propaga la grup
3. Ensemble NVIDIA (super-120b + nano-9b, PRD #9): acord unanim -> high;
2. Grupare pe similaritate (rapidfuzz token_sort_ratio, threshold conservator):
LLM eticheteaza doar reprezentantul, codul se propaga la grup
3. Ensemble NVIDIA (super-120b + nano-9b): acord unanim -> high;
dezacord (orice divergenta) -> needs_mapping. Vot pe coduri, nu pe
self-confidence. ultra-550b EXCLUS (4-5x mai lent, zero castig)
4. Scrub PII (F3): integrat in or_common.call() (regex nr inmatriculare/VIN)
4. Scrub PII: integrat in or_common.call() (regex nr inmatriculare/VIN)
5. Resumabil: scrie *-partial.json incremental, reia de unde a ramas;
retry/backoff pe 429 gestionat de or_common.call()
6. Output: {denumire, cod, sursa, confidence, grup_rep}
NUL = ancore negativa + supresie, NU promovat la cod RAR (#4)
NUL = ancore negativa + supresie, NU promovat la cod RAR
CLI: python3 tools/mapare-llm/or_label.py [N] [--out path] [--partial path]
[--threshold 85] [--batch 20] [--pace 4.0]
@@ -28,7 +28,7 @@ sys.path.insert(0, os.path.dirname(os.path.abspath(__file__)))
import or_common as oc
from rapidfuzz import fuzz
# Modele NVIDIA (decizie PRD #9: pastram super-120b + nano-9b; aruncam ultra-550b)
# Modele NVIDIA: pastram super-120b + nano-9b; aruncam ultra-550b
MODELS = [
"nvidia/nemotron-3-super-120b-a12b:free",
"nvidia/nemotron-nano-9b-v2:free",
@@ -85,8 +85,8 @@ def ensemble_vote(votes):
- Toate N modele cu acelasi cod valid -> (cod, "high", "ensemble-unanim")
- Toate N modele cu "NUL" -> ("NUL", "high", "ensemble-unanim-nul")
- Orice divergenta / parse-fail partial -> ("?", "needs_mapping", "ensemble-dezacord")
Vot pe coduri, NU pe self-confidence (PRD #10, Eng-F7).
NUL tratat SEPARAT: ancore negativa, nu e cod RAR (#4).
Vot pe coduri, NU pe self-confidence.
NUL tratat SEPARAT: ancore negativa, nu e cod RAR.
Intoarce: (cod_final, confidence, sursa)
cod_final: cod RAR valid | "NUL" | "?" (needs human review)