merge training system

core/enhanced_reward_system_integration.py

@@ -20,6 +20,7 @@ from datetime import datetime
 from core.enhanced_reward_calculator import EnhancedRewardCalculator, TimeFrame
 from core.timeframe_inference_coordinator import TimeframeInferenceCoordinator
 from core.enhanced_rl_training_adapter import EnhancedRLTrainingAdapter
+from core.unified_training_manager import UnifiedTrainingManager
 
 logger = logging.getLogger(__name__)
 
@@ -60,6 +61,12 @@ class EnhancedRewardSystemIntegration:
             training_system=getattr(orchestrator, 'enhanced_training_system', None)
         )
         
+        # Unified Training Manager (always available)
+        self.unified_training = UnifiedTrainingManager(
+            orchestrator=orchestrator,
+            reward_system=self,
+        )
+        
         # Integration state
         self.is_running = False
         self.integration_stats = {
@@ -83,6 +90,7 @@ class EnhancedRewardSystemIntegration:
             # Start core components
             await self.inference_coordinator.start_coordination()
             await self.training_adapter.start_training_loop()
+            await self.unified_training.start()
             
             # Start price monitoring
             asyncio.create_task(self._price_monitoring_loop())
@@ -107,6 +115,7 @@ class EnhancedRewardSystemIntegration:
             # Stop components
             await self.inference_coordinator.stop_coordination()
             await self.training_adapter.stop_training_loop()
+            await self.unified_training.stop()
             
             self.is_running = False