popov/gogo2
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

ETS integration and UI

Browse Source
This commit is contained in:
Dobromir Popov
2025-07-05 00:33:32 +03:00
parent d260e73f9a
commit 97d9bc97ee
4 changed files with 399 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

74
core/orchestrator.py
View File

@ -2498,7 +2498,7 @@ class TradingOrchestrator:
return False
def get_enhanced_training_stats(self) -> Dict[str, Any]:
"""Get enhanced training system statistics"""
"""Get enhanced training system statistics with orchestrator integration"""
try:
if not self.enhanced_training_system:
return {
@ -2507,10 +2507,82 @@ class TradingOrchestrator:
'error': 'Training system not initialized'
}
# Get base stats from enhanced training system
stats = self.enhanced_training_system.get_training_statistics()
stats['training_enabled'] = self.training_enabled
stats['system_available'] = ENHANCED_TRAINING_AVAILABLE
# Add orchestrator-specific training integration data
stats['orchestrator_integration'] = {
'models_connected': len([m for m in [self.rl_agent, self.cnn_model, self.cob_rl_agent, self.decision_model] if m is not None]),
'cob_integration_active': self.cob_integration is not None,
'decision_fusion_enabled': self.decision_fusion_enabled,
'symbols_tracking': len(self.symbols),
'recent_decisions_count': sum(len(decisions) for decisions in self.recent_decisions.values()),
'model_weights': self.model_weights.copy(),
'realtime_processing': self.realtime_processing
}
# Add model-specific training status from orchestrator
stats['model_training_status'] = {}
model_mappings = {
'dqn': self.rl_agent,
'cnn': self.cnn_model,
'cob_rl': self.cob_rl_agent,
'decision': self.decision_model
}
for model_name, model in model_mappings.items():
if model:
model_stats = {
'model_loaded': True,
'memory_usage': 0,
'training_steps': 0,
'last_loss': None,
'checkpoint_loaded': self.model_states.get(model_name, {}).get('checkpoint_loaded', False)
}
# Get memory usage
if hasattr(model, 'memory') and model.memory:
model_stats['memory_usage'] = len(model.memory)
# Get training steps
if hasattr(model, 'training_steps'):
model_stats['training_steps'] = model.training_steps
# Get last loss
if hasattr(model, 'losses') and model.losses:
model_stats['last_loss'] = model.losses[-1]
stats['model_training_status'][model_name] = model_stats
else:
stats['model_training_status'][model_name] = {
'model_loaded': False,
'memory_usage': 0,
'training_steps': 0,
'last_loss': None,
'checkpoint_loaded': False
}
# Add prediction tracking stats
stats['prediction_tracking'] = {
'dqn_predictions_tracked': sum(len(preds) for preds in self.recent_dqn_predictions.values()),
'cnn_predictions_tracked': sum(len(preds) for preds in self.recent_cnn_predictions.values()),
'accuracy_history_tracked': sum(len(history) for history in self.prediction_accuracy_history.values()),
'symbols_with_predictions': [symbol for symbol in self.symbols if
len(self.recent_dqn_predictions.get(symbol, [])) > 0 or
len(self.recent_cnn_predictions.get(symbol, [])) > 0]
}
# Add COB integration stats if available
if self.cob_integration:
stats['cob_integration_stats'] = {
'latest_cob_data_symbols': list(self.latest_cob_data.keys()),
'cob_features_available': list(self.latest_cob_features.keys()),
'cob_state_available': list(self.latest_cob_state.keys()),
'feature_history_length': {symbol: len(history) for symbol, history in self.cob_feature_history.items()}
}
return stats
except Exception as e: