Pipeline complet Radiacode 103 - identification automatique d'isotopes

- VegaModel CNN-FCNN 34.5M params, 82 isotopes, val acc 99.89%
- Generation 50k spectres synthetiques 1D (12-24h durees)
- Entrainement 100 epochs sur RTX 5060 Ti (CUDA 12.8, Blackwell)
- Detection continue avec soustraction du background
- Capture background 24h avec gestion deconnexion
- Docker Compose : conteneur train (GPU) + detect (CPU/USB)
- Modele entraite inclus (vega_best.pt, 395 Mo)

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

train/vega_ml/synthetic_spectra/__init__.py

@@ -0,0 +1,18 @@
+"""
+Synthetic Gamma Spectra Generation Module
+
+This module provides tools for generating realistic synthetic gamma spectra
+for training isotope identification models. It simulates detector responses
+compatible with Radiacode devices (101, 102, 103, 103G, 110).
+
+Detector Specifications:
+- Energy Range: 20 keV to 3000 keV (0.02 - 3 MeV)
+- Channels: 1024 (usable: 1023)
+- FWHM Resolution: 7.4% - 9.5% @ 662 keV (model dependent)
+- Detector Types: CsI(Tl) and GAGG(Ce) scintillators
+"""
+
+__version__ = "0.1.0"
+__author__ = "Isotope ID ML Project"
+
+from .config import DetectorConfig, RADIACODE_CONFIGS