feat: integrate dotenv for environment variable management and update weight management logic for prediction handling

app_mcp.py

@@ -26,12 +26,17 @@ from models.smart_agents import ContextualIntelligenceAgent, ForensicAnomalyDete
 
 from forensics.registry import register_model, MODEL_REGISTRY, ModelEntry
 from models.weight_management import ModelWeightManager
+from dotenv import load_dotenv
 
 # Configure logging
 logging.basicConfig(level=logging.DEBUG)
 logger = logging.getLogger(__name__)
+os.environ['HF_HUB_CACHE'] = './models'
 
 
+load_dotenv()
+print(os.getenv("HF_HUB_CACHE"))
+
 # Ensure using GPU if available
 device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
 

models/weight_management.py

@@ -87,13 +87,13 @@ class ModelWeightManager:
     def _has_consensus(self, predictions):
         """Check if models agree on prediction"""
         # Ensure all predictions are not None before checking for consensus
-        non_none_predictions = [p for p in predictions.values() if p is not None and p != "Error"]
+        non_none_predictions = [p.get("Label") for p in predictions.values() if p is not None and isinstance(p, dict) and p.get("Label") is not None and p.get("Label") != "Error"]
         return len(non_none_predictions) > 0 and len(set(non_none_predictions)) == 1
     
     def _has_conflicts(self, predictions):
         """Check if models have conflicting predictions"""
         # Ensure all predictions are not None before checking for conflicts
-        non_none_predictions = [p for p in predictions.values() if p is not None and p != "Error"]
+        non_none_predictions = [p.get("Label") for p in predictions.values() if p is not None and isinstance(p, dict) and p.get("Label") is not None and p.get("Label") != "Error"]
         return len(non_none_predictions) > 1 and len(set(non_none_predictions)) > 1
     
     def _normalize_weights(self, weights):

requirements.txt

@@ -15,3 +15,4 @@ pyexiftool
 psutil
 datasets
 Pillow
+python-dotenv