Update app.py

app.py

@@ -5,6 +5,7 @@ import requests
 from PIL import Image
 from difflib import get_close_matches
 from functools import lru_cache
+from collections import Counter
 
 # =====================
 # ENV
@@ -20,18 +21,27 @@ with open("nutrition_db.json", "r") as f:
 print("✅ Loaded DB:", len(NUTRITION_DB))
 
 # =====================
-# NORMALIZATION
+# COUNTABLE FOODS (NEW)
+# =====================
+COUNTABLE_FOODS = {
+    "roti": {"calories": 120, "protein": 3, "carbs": 20, "fat": 2},
+    "bread": {"calories": 80, "protein": 3, "carbs": 15, "fat": 1},
+    "samosa": {"calories": 260, "protein": 5, "carbs": 30, "fat": 14},
+    "gulab jamun": {"calories": 150, "protein": 2, "carbs": 30, "fat": 5},
+    "laddu": {"calories": 180, "protein": 3, "carbs": 25, "fat": 8},
+    "idli": {"calories": 60, "protein": 2, "carbs": 12, "fat": 1},
+    "vada": {"calories": 150, "protein": 4, "carbs": 20, "fat": 8}
+}
+
+# =====================
+# NORMALIZATION (FIXED)
 # =====================
 def normalize_food_name(name):
     name = name.lower()
 
     mapping = {
-        "chapati": "wheat",
-        "roti": "wheat",
-        "naan": "wheat",
-        "paratha": "wheat",
-        "omelette": "egg",
-        "omellete": "egg",
+        "chapati": "roti",
+        "omellete": "omelette",
         "fried rice": "rice",
         "plain rice": "rice"
     }
@@ -87,7 +97,7 @@ def get_nutrition(dish, grams):
     return compute_nutrition_cached(food_key, grams)
 
 # =====================
-# QUANTITY
+# QUANTITY ESTIMATION
 # =====================
 def estimate_quantity(pred):
     width = pred.get("width", 0)
@@ -100,53 +110,42 @@ def estimate_quantity(pred):
     return round(grams, 1)
 
 # =====================
-# ROBOFLOW (HTTP)
+# ROBOFLOW DETECTION
 # =====================
-
-
-import requests
-import os
-
 def detect(image_path):
     api_key = os.getenv("ROBOFLOW_API_KEY")
 
     if not api_key:
         return "❌ API KEY NOT FOUND"
 
-    # model_id = "almost-final/1"
     url = "https://detect.roboflow.com/almost-final/2"
 
     try:
         with open(image_path, "rb") as f:
             response = requests.post(
                 url,
-                files={"file": f},   # ✅ correct format
+                files={"file": f},
                 params={"api_key": api_key},
                 timeout=15
             )
 
-        print("STATUS:", response.status_code)
-        print("RESPONSE:", response.text)
-
         if response.status_code != 200:
             return f"❌ Roboflow Error: {response.text}"
 
         data = response.json()
-
         return data.get("predictions", [])
 
     except Exception as e:
         return f"❌ Request Failed: {str(e)}"
+
 # =====================
-# MAIN FUNCTION
+# MAIN FUNCTION (HYBRID)
 # =====================
 def analyze_image(image):
     if image is None:
         return "Please upload an image"
 
     path = "temp.jpg"
-
-    # PIL image save
     image.save(path)
 
     preds = detect(path)
@@ -157,17 +156,53 @@ def analyze_image(image):
     if len(preds) == 0:
         return "❌ No food detected"
 
+    # GROUP ITEMS
+    dish_counts = Counter([
+        normalize_food_name(p.get("class", "unknown")) for p in preds
+    ])
+
     output = ""
     total = {"calories": 0, "protein": 0, "carbs": 0, "fat": 0}
 
-    for pred in preds:
-        dish = pred.get("class", "unknown")
+    for dish, count in dish_counts.items():
+
+        if dish == "unknown":
+            continue
+
+        # =========================
+        # COUNTABLE FOODS
+        # =========================
+        if dish in COUNTABLE_FOODS:
+            base = COUNTABLE_FOODS[dish]
+
+            nutrition = {
+                "calories": base["calories"] * count,
+                "protein": base["protein"] * count,
+                "carbs": base["carbs"] * count,
+                "fat": base["fat"] * count,
+            }
+
+            output += f"🍽️ {dish} ({count} pcs)\n"
+
+        # =========================
+        # WEIGHT-BASED FOODS
+        # =========================
+        else:
+            relevant_preds = [
+                p for p in preds if normalize_food_name(p.get("class")) == dish
+            ]
+
+            grams_list = [estimate_quantity(p) for p in relevant_preds]
+            grams = sum(grams_list)
+
+            nutrition = get_nutrition(dish, grams)
 
-        grams = estimate_quantity(pred)
-        nutrition = get_nutrition(dish, grams)
+            output += f"🍽️ {dish}\n"
+            output += f"📏 {grams:.1f} g\n"
 
-        output += f"🍽️ {dish}\n"
-        output += f"📏 {grams} g\n"
+        # =========================
+        # OUTPUT
+        # =========================
         output += f"🔥 {nutrition['calories']} kcal\n"
         output += f"💪 Protein: {nutrition['protein']} g\n"
         output += f"🍞 Carbs: {nutrition['carbs']} g\n"
@@ -177,6 +212,7 @@ def analyze_image(image):
         for k in total:
             total[k] += nutrition[k]
 
+    # TOTAL
     output += "\n🧾 TOTAL:\n"
     output += f"🔥 Calories: {round(total['calories'],2)}\n"
     output += f"💪 Protein: {round(total['protein'],2)} g\n"
@@ -190,7 +226,7 @@ def analyze_image(image):
 # =====================
 demo = gr.Interface(
     fn=analyze_image,
-    inputs=gr.Image(type="pil"),   # ✅ FIXED
+    inputs=gr.Image(type="pil"),
     outputs="text",
     title="🍽️ AI Nutritionist",
     description="Upload food image to get calories & macros"