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/training/vega/__init__.py

@@ -0,0 +1,26 @@
+"""
+Vega Model - CNN-FCNN with Multi-Task Heads for Gamma Spectrum Isotope Identification
+
+Architecture based on research findings from:
+- Wang et al. (2026): CNN-FCNN achieves 99.8% accuracy
+- Galib et al. (2021): Hybrid CNN outperforms pure architectures
+- Turner et al. (2021): 1D CNN robust to gain shifts and shielding
+
+Features:
+- 1D CNN backbone for spectral feature extraction
+- Multi-task heads for isotope classification + activity regression
+- Support for 82 isotopes from the synthetic spectra database
+"""
+
+from .model import VegaModel, VegaConfig
+from .dataset import SpectrumDataset, create_data_loaders
+from .train import train_vega, VegaTrainer
+
+__all__ = [
+    'VegaModel',
+    'VegaConfig', 
+    'SpectrumDataset',
+    'create_data_loaders',
+    'train_vega',
+    'VegaTrainer'
+]