Answer with cooking plan and save edamam API tokens

2026-07-27 14:37:05 +02:00
parent 6a42e4a772
commit 81de985580
3 changed files with 1006 additions and 3 deletions
@@ -0,0 +1,98 @@
+from typing import Dict, Any
+from APIs.edamam import EdamamAPI
+from microwaveDishAnalyzer import MicrowaveDishAnalyzer
+from microwaveThermalEngine import MicrowaveThermalEngine, DishThermalState
+
+
+class MicrowaveCookPlanner:
+    """Orchestrates Edamam API, Dish Analyzer, and Thermal Engine into a single workflow."""
+
+    def __init__(self, cm_per_pixel: float = 0.05):
+        self.edamam_api = EdamamAPI()
+        self.analyzer = MicrowaveDishAnalyzer(cm_per_pixel=cm_per_pixel)
+        self.engine = MicrowaveThermalEngine()
+
+    def _extract_edamam_data(self, edamam_resp: Dict[str, Any]) -> tuple[str, float, Dict[str, float]]:
+        """Parses Edamam Vision response to extract label, total mass, and macronutrient grams."""
+        recipe = edamam_resp.get("combined", {}).get("recipe", {})
+        
+        # Fallback to first dish if 'combined' is empty
+        if not recipe and edamam_resp.get("dishes"):
+            recipe = edamam_resp["dishes"][0].get("recipe", {})
+
+        label = recipe.get("label", "Unknown Dish")
+        total_weight = float(recipe.get("totalWeight", 300.0))  # Default 300g fallback
+
+        nutrients = recipe.get("totalNutrients", {})
+        
+        # Extract macronutrients in grams (Edamam nutrient codes)
+        fat_g = float(nutrients.get("FAT", {}).get("quantity", 0.0))
+        protein_g = float(nutrients.get("PROCNT", {}).get("quantity", 0.0))
+        carbs_g = float(nutrients.get("CHOCDF", {}).get("quantity", 0.0))
+        
+        # Water is sometimes omitted in Edamam; infer remaining mass as water if missing
+        if "WATER" in nutrients:
+            water_g = float(nutrients["WATER"].get("quantity", 0.0))
+        else:
+            water_g = max(0.0, total_weight - (fat_g + protein_g + carbs_g))
+
+        macros = {
+            "water_g": water_g,
+            "fat_g": fat_g,
+            "protein_g": protein_g,
+            "carbs_g": carbs_g,
+        }
+
+        return label, total_weight, macros
+
+    def generate_plan(
+        self,
+        image_path: str,
+        height_cm: float,
+        initial_temp_c: float,
+        microwave_wattage: int = 900
+    ) -> Dict[str, Any]:
+        """Main pipeline call to parse an image and return cooking parameters."""
+        
+        # 1. Vision & Nutrient Analysis
+        edamam_resp = self.edamam_api.analyze_dish_image(image_path)
+        food_label, edamam_mass_g, macros = self._extract_edamam_data(edamam_resp)
+
+        # 2. Geometric Volume Calculation
+        vol_data = self.analyzer.estimate_volume(
+            image_path=image_path,
+            height_cm=height_cm,
+            food_label=food_label
+        )
+
+        # 3. Mass Cross-Validation & Density Check
+        mass_data = self.analyzer.reconcile_mass(
+            edamam_mass_g=edamam_mass_g,
+            volume_cm3=vol_data["volume_cm3"],
+            food_label=food_label
+        )
+        final_mass_g = mass_data["final_mass_g"]
+
+        # 4. Thermal State Creation
+        thermal_state = DishThermalState(
+            food_name=food_label,
+            macronutrients=macros,
+            estimated_mass_g=final_mass_g,
+            initial_temp_c=initial_temp_c,
+            volume_cm3=vol_data["volume_cm3"]
+        )
+
+        # 5. Cook Plan Calculation
+        cook_plan = self.engine.calculate_cook_plan(
+            state=thermal_state,
+            microwave_wattage=microwave_wattage
+        )
+
+        # Return consolidated output
+        return {
+            "dish_name": food_label,
+            "reconciled_mass_g": final_mass_g,
+            "mass_validation_status": mass_data["status"],
+            "cook_plan": cook_plan,
+            "geometry": vol_data
+        }