t2a_v2

Dom/t2a_v2

Go to file

dom 44118f69aa fix: SentenceTransformer meta tensor avec accelerate + torch 2.10

low_cpu_mem_usage=False évite les meta tensors lors du chargement
de l'embedding (sentence-transformers 5.x + accelerate 1.12).

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

2026-02-18 01:16:01 +01:00

data

fix: filtre DAS=DP + correction D55.9→D64.9 + enrichissement supplements CIM-10

2026-02-13 14:03:10 +01:00

scripts

feat: mode Validation DIM dans le viewer Flask

2026-02-17 21:43:02 +01:00

src

fix: SentenceTransformer meta tensor avec accelerate + torch 2.10

2026-02-18 01:16:01 +01:00

tests

feat: enrichissement contre-argumentation CPAM — libellés CIM-10, RAG ciblé, reprocess complet

2026-02-17 23:24:10 +01:00

.env.example

feat: configuration externalisée via .env + audit requirements

2026-02-13 19:46:33 +01:00

.gitignore

feat: configuration externalisée via .env + audit requirements

2026-02-13 19:46:33 +01:00

analyze_pdfs.py

feat: pipeline T2A - anonymisation, extraction CIM-10 et intégration edsnlp

2026-02-10 15:24:12 +01:00

rapport_analyse_pdfs.md

feat: pipeline T2A - anonymisation, extraction CIM-10 et intégration edsnlp

2026-02-10 15:24:12 +01:00

requirements.txt

feat: configuration externalisée via .env + audit requirements

2026-02-13 19:46:33 +01:00

run.sh

feat: configuration externalisée via .env + audit requirements

2026-02-13 19:46:33 +01:00