Add POI feature and imporve the overall backend implementation

services/training/app/ml/data_processor.py

@@ -116,20 +116,22 @@ class EnhancedBakeryDataProcessor:
                                    weather_data: pd.DataFrame,
                                    traffic_data: pd.DataFrame,
                                    inventory_product_id: str,
+                                   poi_features: Dict[str, Any] = None,
                                    tenant_id: str = None,
                                    job_id: str = None,
                                    session=None) -> pd.DataFrame:
         """
         Prepare comprehensive training data for a specific product with repository logging.
-        
+
         Args:
             sales_data: Historical sales data for the product
             weather_data: Weather data
             traffic_data: Traffic data
             inventory_product_id: Inventory product UUID for logging
+            poi_features: POI features (location-based, static)
             tenant_id: Optional tenant ID for tracking
             job_id: Optional job ID for tracking
-            
+
         Returns:
             DataFrame ready for Prophet training with 'ds' and 'y' columns plus features
         """
@@ -250,6 +252,18 @@ class EnhancedBakeryDataProcessor:
                         inventory_product_id=inventory_product_id,
                         total_features=len(daily_sales.columns))
 
+            logger.debug("Starting Step 8b: Add POI features",
+                        inventory_product_id=inventory_product_id)
+            # Step 8b: Add POI features (static, location-based)
+            if poi_features:
+                daily_sales = self._add_poi_features(daily_sales, poi_features)
+                logger.debug("Step 8b completed: Add POI features",
+                            inventory_product_id=inventory_product_id,
+                            poi_feature_count=len(poi_features))
+            else:
+                logger.debug("Step 8b skipped: No POI features available",
+                            inventory_product_id=inventory_product_id)
+
             logger.debug("Starting Step 9: Handle missing values",
                         inventory_product_id=inventory_product_id)
             # Step 7: Handle missing values
@@ -331,6 +345,7 @@ class EnhancedBakeryDataProcessor:
                                          future_dates: pd.DatetimeIndex,
                                          weather_forecast: pd.DataFrame = None,
                                          traffic_forecast: pd.DataFrame = None,
+                                         poi_features: Dict[str, Any] = None,
                                          historical_data: pd.DataFrame = None) -> pd.DataFrame:
         """
         Create features for future predictions with proper date handling.
@@ -339,6 +354,7 @@ class EnhancedBakeryDataProcessor:
             future_dates: Future dates to predict
             weather_forecast: Weather forecast data
             traffic_forecast: Traffic forecast data
+            poi_features: POI features (location-based, static)
             historical_data: Historical data for creating lagged and rolling features
 
         Returns:
@@ -390,6 +406,10 @@ class EnhancedBakeryDataProcessor:
                 logger.warning("No historical data provided, lagged features will be NaN")
                 future_df = self._add_advanced_features(future_df)
 
+            # Add POI features (static, location-based)
+            if poi_features:
+                future_df = self._add_poi_features(future_df, poi_features)
+
             future_df = future_df.rename(columns={'date': 'ds'})
 
             # Handle missing values in future data
@@ -1171,7 +1191,42 @@ class EnhancedBakeryDataProcessor:
                 df[col] = df[col].fillna(default_value)
         
         return df
-    
+
+    def _add_poi_features(self, df: pd.DataFrame, poi_features: Dict[str, Any]) -> pd.DataFrame:
+        """
+        Add POI features to training dataframe.
+
+        POI features are static (location-based, not time-varying),
+        so they're broadcast to all rows in the dataframe.
+
+        Args:
+            df: Training dataframe
+            poi_features: Dictionary of POI ML features
+
+        Returns:
+            Dataframe with POI features added as columns
+        """
+        if not poi_features:
+            logger.warning("No POI features to add")
+            return df
+
+        logger.info(f"Adding {len(poi_features)} POI features to dataframe")
+
+        # Add each POI feature as a column with constant value
+        for feature_name, feature_value in poi_features.items():
+            # Convert boolean to int for ML compatibility
+            if isinstance(feature_value, bool):
+                feature_value = 1 if feature_value else 0
+            df[feature_name] = feature_value
+
+        logger.info(
+            "POI features added successfully",
+            feature_count=len(poi_features),
+            feature_names=list(poi_features.keys())[:5]  # Log first 5 for brevity
+        )
+
+        return df
+
     def _prepare_prophet_format(self, df: pd.DataFrame) -> pd.DataFrame:
         """Prepare data in Prophet format with enhanced validation"""
         prophet_df = df.copy()