app adjust

README.md

@@ -1,66 +1,135 @@
 ---
-title: AI Food Scanner
+title: Food Recognition API
 emoji: 🍽️
 colorFrom: yellow
 colorTo: red
-sdk: gradio
-sdk_version: 4.44.0
-app_file: app.py
+sdk: docker
+app_port: 7860
 pinned: false
 license: mit
 tags:
   - food-recognition
   - computer-vision
   - nutrition
-  - ai
+  - fastapi
   - food-101
-  - efficientnet
-  - gradio
+  - pytorch
 ---
 
-# 🍽️ AI Food Scanner
+# 🍽️ Food Recognition API
 
-**Automatic food recognition powered by AI** - Upload an image and get instant food identification with nutritional information!
+**FastAPI backend for AI-powered food recognition** - Accurate classification of 101 food categories with nutritional information.
 
 ## 🎯 Features
 
-- 🤖 **AI-Powered Recognition** - EfficientNet-B0 trained on Food-101 dataset
-- 📊 **101 Food Categories** - From pizza to sushi, desserts to salads
-- 🥗 **Nutritional Information** - Calories, protein, carbs, and fat per 100g
-- 📸 **Image Quality Analysis** - Automatic assessment of upload quality
-- ⚡ **Fast Inference** - <0.5s on GPU, ~2s on CPU
-- 🎨 **Modern UI** - Clean Gradio interface with intuitive controls
+- 🤖 **Food-101 Model** - Pre-trained on 101,000 images
+- 📊 **101 Food Categories** - Pizza, Sushi, Steak, and more
+- 🥗 **Nutritional Data** - Calories, protein, carbs, fat
+- ⚡ **Fast API** - RESTful endpoint with CORS
+- 🔥 **High Accuracy** - ~85% on Food-101 test set
+- 🌐 **Next.js Ready** - Easy integration with frontend
 
-## 🚀 How to Use
+## 🚀 API Endpoint
 
-1. **Upload an image** - Drag & drop or click to select
-2. **Click "Analyze Food"** - AI processes your image
-3. **View Results:**
-   - Primary food match with confidence score
-   - Top 5 alternative predictions
-   - Nutritional breakdown (per 100g)
-   - Image quality metrics
-   - Model information
+### POST `/api/analyze-food`
+
+Analyze a food image and get classification results.
+
+**Request:**
+```bash
+curl -X POST "https://huggingface.co/spaces/YOUR_USERNAME/foodrecognitionapi/api/analyze-food" \
+  -F "file=@pizza.jpg"
+```
+
+**Response:**
+```json
+{
+  "success": true,
+  "primary_prediction": {
+    "label": "pizza",
+    "name": "Pizza",
+    "confidence": 0.94
+  },
+  "top_predictions": [
+    {"label": "pizza", "name": "Pizza", "confidence": 0.94},
+    {"label": "lasagna", "name": "Lasagna", "confidence": 0.03},
+    ...
+  ],
+  "nutrition": {
+    "food_name": "Pizza",
+    "calories": 266,
+    "protein": 11,
+    "carbs": 33,
+    "fat": 10
+  },
+  "model_info": {
+    "model": "nateraw/food",
+    "dataset": "Food-101",
+    "num_classes": 101,
+    "device": "CPU"
+  }
+}
+```
+
+## 📖 Other Endpoints
+
+- **GET `/`** - API info
+- **GET `/health`** - Health check
+- **GET `/docs`** - Interactive API documentation (Swagger)
+
+## 🔧 Next.js Integration
+
+```typescript
+// app/api/analyze-food/route.ts
+export async function POST(request: Request) {
+  const formData = await request.formData();
+
+  const response = await fetch(
+    'https://huggingface.co/spaces/YOUR_USERNAME/foodrecognitionapi/api/analyze-food',
+    {
+      method: 'POST',
+      body: formData,
+    }
+  );
+
+  return Response.json(await response.json());
+}
+```
+
+```typescript
+// Frontend usage
+const analyzeFood = async (file: File) => {
+  const formData = new FormData();
+  formData.append('file', file);
+
+  const res = await fetch('/api/analyze-food', {
+    method: 'POST',
+    body: formData,
+  });
+
+  const data = await res.json();
+  console.log(data.primary_prediction.name); // "Pizza"
+};
+```
 
 ## 📊 Supported Categories
 
 The model recognizes **101 food categories** including:
 
-- **Main Courses:** Pizza, Sushi, Ramen, Steak, Hamburger, etc.
-- **Desserts:** Cheesecake, Ice Cream, Tiramisu, Donuts, etc.
+- **Main Courses:** Pizza, Sushi, Ramen, Steak, Hamburger, Lasagna, Tacos, etc.
+- **Desserts:** Cheesecake, Ice Cream, Tiramisu, Donuts, Chocolate Cake, etc.
 - **Salads:** Caesar Salad, Greek Salad, Caprese Salad, etc.
-- **Appetizers:** Falafel, Hummus, Spring Rolls, Bruschetta, etc.
-- **Fast Food:** French Fries, Hot Dogs, Nachos, Burgers, etc.
+- **Fast Food:** French Fries, Hot Dogs, Nachos, Chicken Wings, etc.
 
-[See full category list →](https://github.com/stratospark/food-101)
+[See full list →](https://github.com/stratospark/food-101)
 
 ## 🔬 Technical Details
 
 ### Model
-- **Architecture:** EfficientNet-B0
-- **Training Dataset:** Food-101 (101,000 images across 101 categories)
-- **Accuracy:** ~85-90% on validation set
-- **Framework:** PyTorch + Hugging Face Transformers
+- **Architecture:** ViT (Vision Transformer)
+- **Training Dataset:** Food-101 (101,000 images)
+- **Accuracy:** ~85% on test set
+- **Model ID:** `nateraw/food`
 
 ### Performance
 | Device | Inference Time |
@@ -69,38 +138,36 @@ The model recognizes **101 food categories** including:
 | CPU (4 cores) | ~2-3s |
 
 ### Stack
-- **Backend:** Python 3.11
-- **UI:** Gradio 4.0
-- **Deep Learning:** PyTorch 2.0+
-- **Deployment:** Hugging Face Spaces
+- **Framework:** FastAPI
+- **ML:** PyTorch + Transformers
+- **Deployment:** Hugging Face Spaces (Docker)
 
 ## 💡 Tips for Best Results
 
-### ✅ Good Images:
+✅ **Good Images:**
 - Well-lit, focused photos
 - Food fills most of the frame
-- Clear, unobstructed view
-- Single dish per image
+- Clear view of the dish
+- Single item per image
 
-### ❌ Avoid:
-- Dark, blurry, or low-quality images
-- Multiple different foods in one image
-- Extreme close-ups
-- Images smaller than 200x200px
+❌ **Avoid:**
+- Dark or blurry images
+- Multiple different foods
+- Extreme angles
+- Very small images (<200px)
 
-## 🔧 Local Development
+## 🛠️ Local Development
 
-### Requirements
 ```bash
+# Install dependencies
 pip install -r requirements.txt
-```
 
-### Run
-```bash
+# Run server
 python app.py
-```
 
-Then open http://localhost:7860 in your browser.
+# Server will start on http://localhost:7860
+# API docs at http://localhost:7860/docs
+```
 
 ## 📝 License
 
@@ -110,17 +177,14 @@ Then open http://localhost:7860 in your browser.
 
 ## ⚠️ Disclaimer
 
-Nutritional information is estimated based on typical values for each food category. For precise nutritional data, consult product packaging or a registered dietitian.
+Nutritional information is estimated based on typical values. For precise data, consult product packaging or a registered dietitian.
 
 ## 🤝 Credits
 
-- Model: [Kaludi/food-category-classification-v2.0](https://huggingface.co/Kaludi/food-category-classification-v2.0)
-- Dataset: [Food-101](https://data.vision.ee.ethz.ch/cvl/datasets_extra/food-101/)
-- Framework: [Hugging Face Transformers](https://huggingface.co/transformers)
-- UI: [Gradio](https://gradio.app)
+- **Model:** [nateraw/food](https://huggingface.co/nateraw/food)
+- **Dataset:** [Food-101](https://data.vision.ee.ethz.ch/cvl/datasets_extra/food-101/)
+- **Framework:** [FastAPI](https://fastapi.tiangolo.com/) + [Transformers](https://huggingface.co/transformers)
 
 ---
 
-**Made with ❤️ using PyTorch, Transformers, and Gradio**
-
-[Report an issue](https://github.com/YOUR_USERNAME/YOUR_REPO/issues) | [View source code](https://github.com/YOUR_USERNAME/YOUR_REPO)
+**Made with ❤️ using PyTorch and FastAPI**

app.py

@@ -1,323 +1,149 @@
 #!/usr/bin/env python3
 """
-🍽️ AI Food Scanner - Production-Ready System
-================================================
-
-Produkciono spreman AI sistem za skeniranje hrane sa Gradio interfejsom.
-
-Ključne karakteristike:
-- ✅ Optimizovan za Hugging Face Spaces (free tier - CPU/T4 GPU)
-- ✅ EfficientNet-B0 model - najbolji balans brzine i tačnosti
-- ✅ Food-101 dataset klasifikacija (101 kategorija hrane)
-- ✅ Moderan Gradio UI sa prikazom rezultata
-- ✅ Detaljne nutritivne informacije
-- ✅ Analiza kvaliteta slike
-- ✅ Sve lokalno - bez vanjskih API ključeva
-
-Model: EfficientNet-B0 pretrained on Food-101
-Tačnost: ~85-90% na Food-101 datasetu
-Brzina: <2 sekunde po slici na CPU, <0.5s na GPU
-
-Autor: AI Assistant
-Verzija: 1.0.0
+🍽️ Food Recognition API - Production Ready
+============================================
+
+FastAPI backend za food recognition optimizovan za Hugging Face Spaces.
+- Koristi PRAVI Food-101 pretrained model
+- REST API endpoint: POST /api/analyze-food
+- CORS enabled za Next.js integraciju
+- 101 kategorija hrane sa visokom tačnošću
+
+Model: nateraw/food (Food-101 dataset - 101 classes)
+Accuracy: ~85% na Food-101 test set
 """
 
 import os
 import logging
-from typing import Dict, Any, Tuple, Optional
-from pathlib import Path
+from typing import Dict, Any, List, Optional
+from io import BytesIO
 
-import gradio as gr
-import numpy as np
-from PIL import Image, ImageEnhance
 import torch
 import torch.nn.functional as F
-from transformers import AutoFeatureExtractor, AutoModelForImageClassification
+from PIL import Image, ImageEnhance
+import numpy as np
+
+from fastapi import FastAPI, File, UploadFile, HTTPException
+from fastapi.middleware.cors import CORSMiddleware
+from fastapi.responses import JSONResponse
+import uvicorn
 
-# ==================== SETUP LOGGING ====================
+from transformers import AutoImageProcessor, AutoModelForImageClassification
+
+# ==================== LOGGING ====================
 logging.basicConfig(
     level=logging.INFO,
-    format='%(asctime)s - %(name)s - %(levelname)s - %(message)s'
+    format='%(asctime)s - %(levelname)s - %(message)s'
 )
 logger = logging.getLogger(__name__)
 
-
 # ==================== FOOD-101 CATEGORIES ====================
-# Kompletna lista kategorija iz Food-101 dataseta
-FOOD_CATEGORIES = [
-    "apple_pie", "baby_back_ribs", "baklava", "beef_carpaccio", "beef_tartare",
-    "beet_salad", "beignets", "bibimbap", "bread_pudding", "breakfast_burrito",
-    "bruschetta", "caesar_salad", "cannoli", "caprese_salad", "carrot_cake",
-    "ceviche", "cheese_plate", "cheesecake", "chicken_curry", "chicken_quesadilla",
-    "chicken_wings", "chocolate_cake", "chocolate_mousse", "churros", "clam_chowder",
-    "club_sandwich", "crab_cakes", "creme_brulee", "croque_madame", "cup_cakes",
-    "deviled_eggs", "donuts", "dumplings", "edamame", "eggs_benedict",
-    "escargots", "falafel", "filet_mignon", "fish_and_chips", "foie_gras",
-    "french_fries", "french_onion_soup", "french_toast", "fried_calamari", "fried_rice",
-    "frozen_yogurt", "garlic_bread", "gnocchi", "greek_salad", "grilled_cheese_sandwich",
-    "grilled_salmon", "guacamole", "gyoza", "hamburger", "hot_and_sour_soup",
-    "hot_dog", "huevos_rancheros", "hummus", "ice_cream", "lasagna",
-    "lobster_bisque", "lobster_roll_sandwich", "macaroni_and_cheese", "macarons", "miso_soup",
-    "mussels", "nachos", "omelette", "onion_rings", "oysters",
-    "pad_thai", "paella", "pancakes", "panna_cotta", "peking_duck",
-    "pho", "pizza", "pork_chop", "poutine", "prime_rib",
-    "pulled_pork_sandwich", "ramen", "ravioli", "red_velvet_cake", "risotto",
-    "samosa", "sashimi", "scallops", "seaweed_salad", "shrimp_and_grits",
-    "spaghetti_bolognese", "spaghetti_carbonara", "spring_rolls", "steak", "strawberry_shortcake",
-    "sushi", "tacos", "takoyaki", "tiramisu", "tuna_tartare",
-    "waffles"
-]
-
-# Mapiranje kategorija na čitljive nazive
-FOOD_NAMES = {
-    "apple_pie": "Apple Pie",
-    "baby_back_ribs": "Baby Back Ribs",
-    "baklava": "Baklava",
-    "beef_carpaccio": "Beef Carpaccio",
-    "beef_tartare": "Beef Tartare",
-    "beet_salad": "Beet Salad",
-    "beignets": "Beignets",
-    "bibimbap": "Bibimbap",
-    "bread_pudding": "Bread Pudding",
-    "breakfast_burrito": "Breakfast Burrito",
-    "bruschetta": "Bruschetta",
-    "caesar_salad": "Caesar Salad",
-    "cannoli": "Cannoli",
-    "caprese_salad": "Caprese Salad",
-    "carrot_cake": "Carrot Cake",
-    "ceviche": "Ceviche",
-    "cheese_plate": "Cheese Plate",
-    "cheesecake": "Cheesecake",
-    "chicken_curry": "Chicken Curry",
-    "chicken_quesadilla": "Chicken Quesadilla",
-    "chicken_wings": "Chicken Wings",
-    "chocolate_cake": "Chocolate Cake",
-    "chocolate_mousse": "Chocolate Mousse",
-    "churros": "Churros",
-    "clam_chowder": "Clam Chowder",
-    "club_sandwich": "Club Sandwich",
-    "crab_cakes": "Crab Cakes",
-    "creme_brulee": "Creme Brulee",
-    "croque_madame": "Croque Madame",
-    "cup_cakes": "Cupcakes",
-    "deviled_eggs": "Deviled Eggs",
-    "donuts": "Donuts",
-    "dumplings": "Dumplings",
-    "edamame": "Edamame",
-    "eggs_benedict": "Eggs Benedict",
-    "escargots": "Escargots",
-    "falafel": "Falafel",
-    "filet_mignon": "Filet Mignon",
-    "fish_and_chips": "Fish and Chips",
-    "foie_gras": "Foie Gras",
-    "french_fries": "French Fries",
-    "french_onion_soup": "French Onion Soup",
-    "french_toast": "French Toast",
-    "fried_calamari": "Fried Calamari",
-    "fried_rice": "Fried Rice",
-    "frozen_yogurt": "Frozen Yogurt",
-    "garlic_bread": "Garlic Bread",
-    "gnocchi": "Gnocchi",
-    "greek_salad": "Greek Salad",
-    "grilled_cheese_sandwich": "Grilled Cheese Sandwich",
-    "grilled_salmon": "Grilled Salmon",
-    "guacamole": "Guacamole",
-    "gyoza": "Gyoza",
-    "hamburger": "Hamburger",
-    "hot_and_sour_soup": "Hot and Sour Soup",
-    "hot_dog": "Hot Dog",
-    "huevos_rancheros": "Huevos Rancheros",
-    "hummus": "Hummus",
-    "ice_cream": "Ice Cream",
-    "lasagna": "Lasagna",
-    "lobster_bisque": "Lobster Bisque",
-    "lobster_roll_sandwich": "Lobster Roll Sandwich",
-    "macaroni_and_cheese": "Macaroni and Cheese",
-    "macarons": "Macarons",
-    "miso_soup": "Miso Soup",
-    "mussels": "Mussels",
-    "nachos": "Nachos",
-    "omelette": "Omelette",
-    "onion_rings": "Onion Rings",
-    "oysters": "Oysters",
-    "pad_thai": "Pad Thai",
-    "paella": "Paella",
-    "pancakes": "Pancakes",
-    "panna_cotta": "Panna Cotta",
-    "peking_duck": "Peking Duck",
-    "pho": "Pho",
-    "pizza": "Pizza",
-    "pork_chop": "Pork Chop",
-    "poutine": "Poutine",
-    "prime_rib": "Prime Rib",
-    "pulled_pork_sandwich": "Pulled Pork Sandwich",
-    "ramen": "Ramen",
-    "ravioli": "Ravioli",
-    "red_velvet_cake": "Red Velvet Cake",
-    "risotto": "Risotto",
-    "samosa": "Samosa",
-    "sashimi": "Sashimi",
-    "scallops": "Scallops",
-    "seaweed_salad": "Seaweed Salad",
-    "shrimp_and_grits": "Shrimp and Grits",
-    "spaghetti_bolognese": "Spaghetti Bolognese",
-    "spaghetti_carbonara": "Spaghetti Carbonara",
-    "spring_rolls": "Spring Rolls",
-    "steak": "Steak",
-    "strawberry_shortcake": "Strawberry Shortcake",
-    "sushi": "Sushi",
-    "tacos": "Tacos",
-    "takoyaki": "Takoyaki",
-    "tiramisu": "Tiramisu",
-    "tuna_tartare": "Tuna Tartare",
-    "waffles": "Waffles"
+FOOD_CATEGORIES = {
+    0: "apple_pie", 1: "baby_back_ribs", 2: "baklava", 3: "beef_carpaccio", 4: "beef_tartare",
+    5: "beet_salad", 6: "beignets", 7: "bibimbap", 8: "bread_pudding", 9: "breakfast_burrito",
+    10: "bruschetta", 11: "caesar_salad", 12: "cannoli", 13: "caprese_salad", 14: "carrot_cake",
+    15: "ceviche", 16: "cheese_plate", 17: "cheesecake", 18: "chicken_curry", 19: "chicken_quesadilla",
+    20: "chicken_wings", 21: "chocolate_cake", 22: "chocolate_mousse", 23: "churros", 24: "clam_chowder",
+    25: "club_sandwich", 26: "crab_cakes", 27: "creme_brulee", 28: "croque_madame", 29: "cup_cakes",
+    30: "deviled_eggs", 31: "donuts", 32: "dumplings", 33: "edamame", 34: "eggs_benedict",
+    35: "escargots", 36: "falafel", 37: "filet_mignon", 38: "fish_and_chips", 39: "foie_gras",
+    40: "french_fries", 41: "french_onion_soup", 42: "french_toast", 43: "fried_calamari", 44: "fried_rice",
+    45: "frozen_yogurt", 46: "garlic_bread", 47: "gnocchi", 48: "greek_salad", 49: "grilled_cheese_sandwich",
+    50: "grilled_salmon", 51: "guacamole", 52: "gyoza", 53: "hamburger", 54: "hot_and_sour_soup",
+    55: "hot_dog", 56: "huevos_rancheros", 57: "hummus", 58: "ice_cream", 59: "lasagna",
+    60: "lobster_bisque", 61: "lobster_roll_sandwich", 62: "macaroni_and_cheese", 63: "macarons", 64: "miso_soup",
+    65: "mussels", 66: "nachos", 67: "omelette", 68: "onion_rings", 69: "oysters",
+    70: "pad_thai", 71: "paella", 72: "pancakes", 73: "panna_cotta", 74: "peking_duck",
+    75: "pho", 76: "pizza", 77: "pork_chop", 78: "poutine", 79: "prime_rib",
+    80: "pulled_pork_sandwich", 81: "ramen", 82: "ravioli", 83: "red_velvet_cake", 84: "risotto",
+    85: "samosa", 86: "sashimi", 87: "scallops", 88: "seaweed_salad", 89: "shrimp_and_grits",
+    90: "spaghetti_bolognese", 91: "spaghetti_carbonara", 92: "spring_rolls", 93: "steak", 94: "strawberry_shortcake",
+    95: "sushi", 96: "tacos", 97: "takoyaki", 98: "tiramisu", 99: "tuna_tartare", 100: "waffles"
 }
 
-
-# ==================== NUTRITIONAL DATABASE ====================
-# Jednostavna baza nutritivnih informacija po kategorijama hrane
-NUTRITION_DATABASE = {
-    # Deserti
-    "apple_pie": {"calories": 237, "protein": 2, "carbs": 34, "fat": 11, "category": "Dessert"},
-    "baklava": {"calories": 334, "protein": 4, "carbs": 29, "fat": 23, "category": "Dessert"},
-    "cannoli": {"calories": 213, "protein": 5, "carbs": 25, "fat": 11, "category": "Dessert"},
-    "carrot_cake": {"calories": 415, "protein": 4, "carbs": 51, "fat": 21, "category": "Dessert"},
-    "cheesecake": {"calories": 321, "protein": 5, "carbs": 26, "fat": 23, "category": "Dessert"},
-    "chocolate_cake": {"calories": 352, "protein": 4, "carbs": 51, "fat": 16, "category": "Dessert"},
-    "chocolate_mousse": {"calories": 214, "protein": 4, "carbs": 23, "fat": 13, "category": "Dessert"},
-    "churros": {"calories": 237, "protein": 3, "carbs": 29, "fat": 12, "category": "Dessert"},
-    "creme_brulee": {"calories": 297, "protein": 4, "carbs": 26, "fat": 20, "category": "Dessert"},
-    "cup_cakes": {"calories": 305, "protein": 4, "carbs": 45, "fat": 13, "category": "Dessert"},
-    "donuts": {"calories": 269, "protein": 3, "carbs": 31, "fat": 15, "category": "Dessert"},
-    "ice_cream": {"calories": 207, "protein": 4, "carbs": 24, "fat": 11, "category": "Dessert"},
-    "macarons": {"calories": 97, "protein": 2, "carbs": 16, "fat": 3, "category": "Dessert"},
-    "panna_cotta": {"calories": 305, "protein": 3, "carbs": 22, "fat": 23, "category": "Dessert"},
-    "red_velvet_cake": {"calories": 380, "protein": 4, "carbs": 53, "fat": 18, "category": "Dessert"},
-    "strawberry_shortcake": {"calories": 247, "protein": 3, "carbs": 38, "fat": 10, "category": "Dessert"},
-    "tiramisu": {"calories": 240, "protein": 5, "carbs": 26, "fat": 13, "category": "Dessert"},
-    "waffles": {"calories": 291, "protein": 6, "carbs": 33, "fat": 15, "category": "Dessert"},
-
-    # Glavni obroci
-    "baby_back_ribs": {"calories": 361, "protein": 27, "carbs": 0, "fat": 27, "category": "Main Course"},
-    "beef_carpaccio": {"calories": 129, "protein": 22, "carbs": 1, "fat": 4, "category": "Main Course"},
-    "beef_tartare": {"calories": 220, "protein": 20, "carbs": 2, "fat": 15, "category": "Main Course"},
-    "bibimbap": {"calories": 560, "protein": 25, "carbs": 80, "fat": 15, "category": "Main Course"},
-    "chicken_curry": {"calories": 288, "protein": 20, "carbs": 15, "fat": 17, "category": "Main Course"},
-    "chicken_quesadilla": {"calories": 529, "protein": 27, "carbs": 39, "fat": 29, "category": "Main Course"},
-    "chicken_wings": {"calories": 203, "protein": 23, "carbs": 0, "fat": 12, "category": "Main Course"},
-    "filet_mignon": {"calories": 227, "protein": 26, "carbs": 0, "fat": 13, "category": "Main Course"},
-    "fish_and_chips": {"calories": 585, "protein": 32, "carbs": 51, "fat": 28, "category": "Main Course"},
-    "grilled_salmon": {"calories": 206, "protein": 22, "carbs": 0, "fat": 12, "category": "Main Course"},
-    "hamburger": {"calories": 354, "protein": 20, "carbs": 30, "fat": 17, "category": "Main Course"},
-    "lasagna": {"calories": 315, "protein": 14, "carbs": 30, "fat": 15, "category": "Main Course"},
-    "pad_thai": {"calories": 429, "protein": 17, "carbs": 61, "fat": 13, "category": "Main Course"},
-    "paella": {"calories": 525, "protein": 28, "carbs": 58, "fat": 19, "category": "Main Course"},
-    "peking_duck": {"calories": 337, "protein": 19, "carbs": 1, "fat": 28, "category": "Main Course"},
-    "pho": {"calories": 350, "protein": 15, "carbs": 45, "fat": 12, "category": "Main Course"},
-    "pizza": {"calories": 266, "protein": 11, "carbs": 33, "fat": 10, "category": "Main Course"},
-    "pork_chop": {"calories": 231, "protein": 27, "carbs": 0, "fat": 13, "category": "Main Course"},
-    "prime_rib": {"calories": 338, "protein": 26, "carbs": 0, "fat": 26, "category": "Main Course"},
-    "ramen": {"calories": 436, "protein": 15, "carbs": 52, "fat": 19, "category": "Main Course"},
-    "risotto": {"calories": 200, "protein": 4, "carbs": 30, "fat": 6, "category": "Main Course"},
-    "spaghetti_bolognese": {"calories": 281, "protein": 14, "carbs": 34, "fat": 10, "category": "Main Course"},
-    "spaghetti_carbonara": {"calories": 311, "protein": 13, "carbs": 36, "fat": 13, "category": "Main Course"},
-    "steak": {"calories": 271, "protein": 26, "carbs": 0, "fat": 18, "category": "Main Course"},
-    "sushi": {"calories": 143, "protein": 6, "carbs": 21, "fat": 4, "category": "Main Course"},
-    "tacos": {"calories": 226, "protein": 9, "carbs": 20, "fat": 13, "category": "Main Course"},
-
-    # Salate i predjela
-    "beet_salad": {"calories": 152, "protein": 4, "carbs": 18, "fat": 8, "category": "Salad"},
-    "caesar_salad": {"calories": 184, "protein": 9, "carbs": 8, "fat": 13, "category": "Salad"},
-    "caprese_salad": {"calories": 286, "protein": 11, "carbs": 6, "fat": 24, "category": "Salad"},
-    "greek_salad": {"calories": 107, "protein": 4, "carbs": 8, "fat": 7, "category": "Salad"},
-    "seaweed_salad": {"calories": 70, "protein": 2, "carbs": 14, "fat": 1, "category": "Salad"},
-    "bruschetta": {"calories": 77, "protein": 2, "carbs": 11, "fat": 3, "category": "Appetizer"},
-    "ceviche": {"calories": 130, "protein": 20, "carbs": 8, "fat": 2, "category": "Appetizer"},
-    "deviled_eggs": {"calories": 145, "protein": 6, "carbs": 1, "fat": 13, "category": "Appetizer"},
-    "edamame": {"calories": 122, "protein": 11, "carbs": 10, "fat": 5, "category": "Appetizer"},
-    "falafel": {"calories": 333, "protein": 13, "carbs": 32, "fat": 18, "category": "Appetizer"},
-    "fried_calamari": {"calories": 175, "protein": 18, "carbs": 8, "fat": 7, "category": "Appetizer"},
-    "guacamole": {"calories": 150, "protein": 2, "carbs": 9, "fat": 13, "category": "Appetizer"},
-    "hummus": {"calories": 166, "protein": 5, "carbs": 14, "fat": 10, "category": "Appetizer"},
-    "spring_rolls": {"calories": 109, "protein": 3, "carbs": 15, "fat": 4, "category": "Appetizer"},
-
-    # Sendviči i brza hrana
-    "breakfast_burrito": {"calories": 653, "protein": 28, "carbs": 60, "fat": 33, "category": "Fast Food"},
-    "club_sandwich": {"calories": 590, "protein": 31, "carbs": 47, "fat": 30, "category": "Fast Food"},
-    "french_fries": {"calories": 312, "protein": 3, "carbs": 37, "fat": 17, "category": "Fast Food"},
-    "grilled_cheese_sandwich": {"calories": 393, "protein": 17, "carbs": 32, "fat": 22, "category": "Fast Food"},
-    "hot_dog": {"calories": 290, "protein": 10, "carbs": 24, "fat": 17, "category": "Fast Food"},
-    "lobster_roll_sandwich": {"calories": 436, "protein": 30, "carbs": 35, "fat": 18, "category": "Fast Food"},
-    "nachos": {"calories": 346, "protein": 9, "carbs": 36, "fat": 19, "category": "Fast Food"},
-    "pulled_pork_sandwich": {"calories": 508, "protein": 29, "carbs": 41, "fat": 23, "category": "Fast Food"},
-
-    # Default za ostale kategorije
-    "default": {"calories": 200, "protein": 10, "carbs": 25, "fat": 8, "category": "Unknown"}
+# Readable names
+FOOD_NAMES = {
+    "apple_pie": "Apple Pie", "baby_back_ribs": "Baby Back Ribs", "baklava": "Baklava",
+    "beef_carpaccio": "Beef Carpaccio", "beef_tartare": "Beef Tartare", "beet_salad": "Beet Salad",
+    "beignets": "Beignets", "bibimbap": "Bibimbap", "bread_pudding": "Bread Pudding",
+    "breakfast_burrito": "Breakfast Burrito", "bruschetta": "Bruschetta", "caesar_salad": "Caesar Salad",
+    "cannoli": "Cannoli", "caprese_salad": "Caprese Salad", "carrot_cake": "Carrot Cake",
+    "ceviche": "Ceviche", "cheese_plate": "Cheese Plate", "cheesecake": "Cheesecake",
+    "chicken_curry": "Chicken Curry", "chicken_quesadilla": "Chicken Quesadilla",
+    "chicken_wings": "Chicken Wings", "chocolate_cake": "Chocolate Cake",
+    "chocolate_mousse": "Chocolate Mousse", "churros": "Churros", "clam_chowder": "Clam Chowder",
+    "club_sandwich": "Club Sandwich", "crab_cakes": "Crab Cakes", "creme_brulee": "Creme Brulee",
+    "croque_madame": "Croque Madame", "cup_cakes": "Cupcakes", "deviled_eggs": "Deviled Eggs",
+    "donuts": "Donuts", "dumplings": "Dumplings", "edamame": "Edamame",
+    "eggs_benedict": "Eggs Benedict", "escargots": "Escargots", "falafel": "Falafel",
+    "filet_mignon": "Filet Mignon", "fish_and_chips": "Fish and Chips", "foie_gras": "Foie Gras",
+    "french_fries": "French Fries", "french_onion_soup": "French Onion Soup",
+    "french_toast": "French Toast", "fried_calamari": "Fried Calamari", "fried_rice": "Fried Rice",
+    "frozen_yogurt": "Frozen Yogurt", "garlic_bread": "Garlic Bread", "gnocchi": "Gnocchi",
+    "greek_salad": "Greek Salad", "grilled_cheese_sandwich": "Grilled Cheese Sandwich",
+    "grilled_salmon": "Grilled Salmon", "guacamole": "Guacamole", "gyoza": "Gyoza",
+    "hamburger": "Hamburger", "hot_and_sour_soup": "Hot and Sour Soup", "hot_dog": "Hot Dog",
+    "huevos_rancheros": "Huevos Rancheros", "hummus": "Hummus", "ice_cream": "Ice Cream",
+    "lasagna": "Lasagna", "lobster_bisque": "Lobster Bisque",
+    "lobster_roll_sandwich": "Lobster Roll Sandwich", "macaroni_and_cheese": "Macaroni and Cheese",
+    "macarons": "Macarons", "miso_soup": "Miso Soup", "mussels": "Mussels", "nachos": "Nachos",
+    "omelette": "Omelette", "onion_rings": "Onion Rings", "oysters": "Oysters",
+    "pad_thai": "Pad Thai", "paella": "Paella", "pancakes": "Pancakes", "panna_cotta": "Panna Cotta",
+    "peking_duck": "Peking Duck", "pho": "Pho", "pizza": "Pizza", "pork_chop": "Pork Chop",
+    "poutine": "Poutine", "prime_rib": "Prime Rib", "pulled_pork_sandwich": "Pulled Pork Sandwich",
+    "ramen": "Ramen", "ravioli": "Ravioli", "red_velvet_cake": "Red Velvet Cake",
+    "risotto": "Risotto", "samosa": "Samosa", "sashimi": "Sashimi", "scallops": "Scallops",
+    "seaweed_salad": "Seaweed Salad", "shrimp_and_grits": "Shrimp and Grits",
+    "spaghetti_bolognese": "Spaghetti Bolognese", "spaghetti_carbonara": "Spaghetti Carbonara",
+    "spring_rolls": "Spring Rolls", "steak": "Steak", "strawberry_shortcake": "Strawberry Shortcake",
+    "sushi": "Sushi", "tacos": "Tacos", "takoyaki": "Takoyaki", "tiramisu": "Tiramisu",
+    "tuna_tartare": "Tuna Tartare", "waffles": "Waffles"
 }
 
+# Nutrition database
+NUTRITION_DB = {
+    "pizza": {"calories": 266, "protein": 11, "carbs": 33, "fat": 10},
+    "hamburger": {"calories": 354, "protein": 20, "carbs": 30, "fat": 17},
+    "sushi": {"calories": 143, "protein": 6, "carbs": 21, "fat": 4},
+    "ice_cream": {"calories": 207, "protein": 4, "carbs": 24, "fat": 11},
+    "french_fries": {"calories": 312, "protein": 3, "carbs": 37, "fat": 17},
+    "chicken_wings": {"calories": 203, "protein": 23, "carbs": 0, "fat": 12},
+    "chocolate_cake": {"calories": 352, "protein": 4, "carbs": 51, "fat": 16},
+    "caesar_salad": {"calories": 184, "protein": 9, "carbs": 8, "fat": 13},
+    "steak": {"calories": 271, "protein": 26, "carbs": 0, "fat": 18},
+    "tacos": {"calories": 226, "protein": 9, "carbs": 20, "fat": 13},
+    # Default for others
+    "_default": {"calories": 200, "protein": 10, "carbs": 25, "fat": 8}
+}
 
 # ==================== DEVICE SELECTION ====================
 def select_device() -> str:
-    """
-    Automatski odabir najboljeg dostupnog device-a.
-    Optimizovano za Hugging Face Spaces (CPU ili T4 GPU).
-
-    Returns:
-        str: 'cuda', 'mps', ili 'cpu'
-    """
-    # Check for CUDA GPU (HF Spaces T4 GPU)
+    """Select best available device."""
     if torch.cuda.is_available():
-        device = "cuda"
-        gpu_name = torch.cuda.get_device_name(0)
-        gpu_memory = torch.cuda.get_device_properties(0).total_memory / (1024**3)
-        logger.info(f"✅ CUDA available - Using GPU: {gpu_name} ({gpu_memory:.1f} GB)")
-    # Check for Apple Silicon MPS (local development)
+        logger.info(f"✅ Using CUDA GPU: {torch.cuda.get_device_name(0)}")
+        return "cuda"
     elif hasattr(torch.backends, "mps") and torch.backends.mps.is_available():
-        device = "mps"
-        logger.info("✅ MPS available - Using Apple Silicon GPU")
-    # Fallback to CPU (HF Spaces free tier default)
+        logger.info("✅ Using Apple Silicon GPU (MPS)")
+        return "mps"
     else:
-        device = "cpu"
-        logger.info("⚠️ Using CPU - inference will be slower (~2-3s per image)")
-
-    return device
-
+        logger.info("⚠️ Using CPU")
+        return "cpu"
 
 # ==================== IMAGE PREPROCESSING ====================
 def preprocess_image(image: Image.Image) -> Image.Image:
-    """
-    Napredna predobrada slike za bolju klasifikaciju.
-
-    Koraci:
-    1. Konverzija u RGB ako nije
-    2. Poboljšanje oštrine
-    3. Poboljšanje kontrasta
-    4. Optimizacija veličine (za memoriju)
-
-    Args:
-        image: PIL Image objekat
-
-    Returns:
-        PIL Image: Predobrađena slika
-    """
-    # Konverzija u RGB
+    """Enhanced image preprocessing."""
     if image.mode != "RGB":
         image = image.convert("RGB")
 
-    # Poboljšanje oštrine
+    # Enhance image
     enhancer = ImageEnhance.Sharpness(image)
-    image = enhancer.enhance(1.3)
-
-    # Poboljšanje kontrasta
-    enhancer = ImageEnhance.Contrast(image)
     image = enhancer.enhance(1.2)
 
-    # Blago poboljšanje boja
-    enhancer = ImageEnhance.Color(image)
-    image = enhancer.enhance(1.1)
+    enhancer = ImageEnhance.Contrast(image)
+    image = enhancer.enhance(1.15)
 
-    # Resize ako je prevelika (za optimizaciju memorije)
-    max_size = 800
+    # Resize if too large
+    max_size = 512
     if max(image.size) > max_size:
         ratio = max_size / max(image.size)
         new_size = tuple(int(dim * ratio) for dim in image.size)
@@ -325,455 +151,191 @@ def preprocess_image(image: Image.Image) -> Image.Image:
 
     return image
 
-
-# ==================== IMAGE QUALITY ANALYSIS ====================
-def analyze_image_quality(image: Image.Image) -> Dict[str, Any]:
-    """
-    Analizira kvalitet slike za detekciju problema.
-
-    Args:
-        image: PIL Image objekat
-
-    Returns:
-        Dict sa metrikama kvaliteta
-    """
-    img_array = np.array(image)
-
-    # Brightness analiza
-    brightness = float(np.mean(img_array))
-
-    # Color saturation analiza
-    r, g, b = img_array[:, :, 0], img_array[:, :, 1], img_array[:, :, 2]
-    saturation = float(np.mean(np.abs(r - g) + np.abs(g - b) + np.abs(b - r)))
-
-    # Texture complexity (variance)
-    texture_complexity = float(np.var(img_array) / 10000)
-
-    # Overall quality score (0-10)
-    quality_score = 5.0
-
-    # Brightness assessment
-    if 80 <= brightness <= 200:
-        quality_score += 2
-    elif brightness < 50 or brightness > 220:
-        quality_score -= 2
-
-    # Saturation assessment
-    if saturation > 30:
-        quality_score += 2
-    elif saturation < 10:
-        quality_score -= 1
-
-    # Texture assessment
-    if texture_complexity > 0.1:
-        quality_score += 1
-
-    quality_score = max(0, min(10, quality_score))
-
-    return {
-        "brightness": brightness,
-        "saturation": saturation,
-        "texture_complexity": texture_complexity,
-        "quality_score": quality_score,
-        "width": image.width,
-        "height": image.height
-    }
-
-
-# ==================== FOOD RECOGNIZER CLASS ====================
+# ==================== FOOD RECOGNIZER ====================
 class FoodRecognizer:
-    """
-    Glavni AI model za prepoznavanje hrane.
-
-    Koristi EfficientNet-B0 pretrained na Food-101 datasetu.
-    - 101 kategorija hrane
-    - ~85-90% tačnost
-    - Optimizovan za CPU i GPU
-    """
+    """Food recognition using Food-101 trained model."""
 
     def __init__(self, device: str):
-        """
-        Inicijalizacija modela.
-
-        Args:
-            device: 'cuda', 'mps', ili 'cpu'
-        """
         self.device = device
         self.model = None
-        self.feature_extractor = None
-
-        logger.info("🔄 Loading AI model...")
+        self.processor = None
         self._load_model()
 
     def _load_model(self):
-        """
-        Učitava EfficientNet-B0 model treniran na Food-101.
-
-        Model se automatski preuzima sa Hugging Face Hub.
-        """
-        # Setup cache directory za HF Spaces
-        cache_dir = os.environ.get("TRANSFORMERS_CACHE", None)
-
+        """Load Food-101 trained model."""
         try:
-            # Model ID - pretrained na Food-101
-            model_name = "Kaludi/food-category-classification-v2.0"
+            # Use nateraw/food - PRAVI Food-101 model
+            model_name = "nateraw/food"
 
-            logger.info(f"📥 Downloading model: {model_name}")
+            logger.info(f"📥 Loading model: {model_name}")
 
-            # Load kwargs sa cache directory
-            load_kwargs = {}
-            if cache_dir:
-                load_kwargs["cache_dir"] = cache_dir
+            # Setup cache
+            cache_dir = os.environ.get("TRANSFORMERS_CACHE", None)
+            load_kwargs = {"cache_dir": cache_dir} if cache_dir else {}
 
-            # Učitaj feature extractor
-            self.feature_extractor = AutoFeatureExtractor.from_pretrained(
-                model_name,
-                **load_kwargs
-            )
-
-            # Učitaj model (force safetensors for security)
+            # Load processor and model
+            self.processor = AutoImageProcessor.from_pretrained(model_name, **load_kwargs)
             self.model = AutoModelForImageClassification.from_pretrained(
                 model_name,
-                torch_dtype=torch.float16 if self.device == "cuda" else torch.float32,
-                use_safetensors=True,  # Force safetensors format (security)
+                use_safetensors=True,
                 **load_kwargs
             )
 
-            # Prebaci na device i postavi u eval mode
             self.model = self.model.to(self.device)
             self.model.eval()
 
-            logger.info(f"✅ Model loaded successfully on {self.device.upper()}")
+            logger.info(f"✅ Model loaded on {self.device.upper()}")
 
         except Exception as e:
             logger.error(f"❌ Failed to load model: {e}")
-            # Fallback na drugi model ako prvi ne radi
-            try:
-                logger.info("🔄 Trying fallback model...")
-                model_name = "nateraw/food"
-
-                load_kwargs = {}
-                if cache_dir:
-                    load_kwargs["cache_dir"] = cache_dir
-
-                self.feature_extractor = AutoFeatureExtractor.from_pretrained(
-                    model_name,
-                    **load_kwargs
-                )
-                self.model = AutoModelForImageClassification.from_pretrained(
-                    model_name,
-                    use_safetensors=True,  # Force safetensors format (security)
-                    **load_kwargs
-                )
-                self.model = self.model.to(self.device)
-                self.model.eval()
-
-                logger.info(f"✅ Fallback model loaded on {self.device.upper()}")
-            except Exception as e2:
-                logger.error(f"❌ Fallback model also failed: {e2}")
-                raise RuntimeError("Unable to load any model")
+            raise RuntimeError(f"Model loading failed: {e}")
 
     def predict(self, image: Image.Image, top_k: int = 5) -> Dict[str, Any]:
-        """
-        Predikcija kategorije hrane iz slike.
-
-        Args:
-            image: PIL Image objekat
-            top_k: Broj top rezultata za vratiti
-
-        Returns:
-            Dict sa rezultatima predikcije
-        """
-        # Predobrada slike
+        """Predict food category from image."""
+        # Preprocess
         processed_image = preprocess_image(image)
 
-        # Analiza kvaliteta
-        quality_metrics = analyze_image_quality(processed_image)
-
-        # Ekstrakcija features-a
-        inputs = self.feature_extractor(
-            images=processed_image,
-            return_tensors="pt"
-        )
-
-        # Prebaci na device
+        # Prepare inputs
+        inputs = self.processor(images=processed_image, return_tensors="pt")
         inputs = {k: v.to(self.device) for k, v in inputs.items()}
 
-        # Predikcija (bez gradijenta)
+        # Inference
         with torch.no_grad():
             outputs = self.model(**inputs)
             logits = outputs.logits
+            probs = F.softmax(logits, dim=-1).cpu().numpy()[0]
 
-            # Softmax za probabilnosti
-            probs = F.softmax(logits, dim=-1)
-            probs = probs.cpu().numpy()[0]
-
-        # Top K rezultata
+        # Get top K predictions
         top_indices = np.argsort(probs)[::-1][:top_k]
 
         results = []
         for idx in top_indices:
-            label = self.model.config.id2label[idx]
+            label_key = self.model.config.id2label[idx]
             confidence = float(probs[idx])
 
-            # Dobij čitljivo ime
-            readable_name = FOOD_NAMES.get(label, label.replace("_", " ").title())
+            # Get readable name
+            readable_name = FOOD_NAMES.get(label_key, label_key.replace("_", " ").title())
 
             results.append({
-                "label": label,
+                "label": label_key,
                 "name": readable_name,
                 "confidence": confidence
             })
 
-        # Glavni rezultat
-        primary_result = results[0]
-
-        # Dobij nutritivne informacije
-        nutrition = self._get_nutrition(primary_result["label"])
+        # Get nutrition info
+        primary_label = results[0]["label"]
+        nutrition = NUTRITION_DB.get(primary_label, NUTRITION_DB["_default"]).copy()
+        nutrition["food_name"] = results[0]["name"]
 
         return {
-            "primary_prediction": primary_result,
+            "success": True,
+            "primary_prediction": results[0],
             "top_predictions": results,
             "nutrition": nutrition,
-            "image_quality": quality_metrics,
             "model_info": {
-                "device": self.device.upper(),
-                "model_type": "EfficientNet-B0",
+                "model": "nateraw/food",
                 "dataset": "Food-101",
-                "num_categories": 101
+                "num_classes": 101,
+                "device": self.device.upper()
             }
         }
 
-    def _get_nutrition(self, food_label: str) -> Dict[str, Any]:
-        """
-        Dobija nutritivne informacije za hranu.
+# ==================== FASTAPI APP ====================
+logger.info("=" * 80)
+logger.info("🍽️ FOOD RECOGNITION API - STARTING")
+logger.info("=" * 80)
 
-        Args:
-            food_label: Oznaka kategorije hrane
+# Initialize model
+device = select_device()
+recognizer = FoodRecognizer(device)
 
-        Returns:
-            Dict sa nutritivnim informacijama
-        """
-        # Ako postoji u bazi, vrati tačne podatke
-        if food_label in NUTRITION_DATABASE:
-            nutrition = NUTRITION_DATABASE[food_label].copy()
-        else:
-            # Inače vrati default estimate
-            nutrition = NUTRITION_DATABASE["default"].copy()
+# Create FastAPI app
+app = FastAPI(
+    title="Food Recognition API",
+    description="AI-powered food recognition with 101 categories",
+    version="1.0.0"
+)
+
+# CORS - enable all origins for Next.js
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=["*"],  # Allow all origins (adjust in production)
+    allow_credentials=True,
+    allow_methods=["*"],
+    allow_headers=["*"],
+)
 
-        # Dodaj readable name
-        nutrition["food_name"] = FOOD_NAMES.get(food_label, food_label.replace("_", " ").title())
+# ==================== API ENDPOINTS ====================
 
-        return nutrition
+@app.get("/")
+def root():
+    """Root endpoint."""
+    return {
+        "message": "Food Recognition API",
+        "status": "online",
+        "endpoints": {
+            "POST /api/analyze-food": "Analyze food image",
+            "GET /health": "Health check"
+        }
+    }
 
+@app.get("/health")
+def health():
+    """Health check endpoint."""
+    return {
+        "status": "healthy",
+        "model_loaded": recognizer.model is not None,
+        "device": device.upper()
+    }
 
-# ==================== GRADIO INTERFACE ====================
-def create_gradio_interface(recognizer: FoodRecognizer) -> gr.Blocks:
+@app.post("/api/analyze-food")
+async def analyze_food(file: UploadFile = File(...)):
     """
-    Kreira moderan Gradio interfejs za AI Food Scanner.
+    Analyze food image.
 
     Args:
-        recognizer: FoodRecognizer instanca
+        file: Image file (JPEG, PNG, WebP)
 
     Returns:
-        Gradio Blocks interfejs
+        JSON with food recognition results
     """
-
-    def predict_food(image):
-        """
-        Wrapper funkcija za Gradio - procesira sliku i vraća rezultate.
-        """
-        if image is None:
-            return None, "⚠️ Please upload an image first!"
-
-        try:
-            # Konvertuj u PIL Image ako već nije
-            if not isinstance(image, Image.Image):
-                image = Image.fromarray(image)
-
-            # Predikcija
-            logger.info("🔍 Processing image...")
-            results = recognizer.predict(image, top_k=5)
-
-            # Formatiraj output tekst
-            primary = results["primary_prediction"]
-            nutrition = results["nutrition"]
-            quality = results["image_quality"]
-
-            output_text = f"""
-# 🍽️ Detection Results
-
-## Primary Match
-**{primary['name']}**
-Confidence: **{primary['confidence']:.1%}**
-
-## Top 5 Predictions
-"""
-            for i, pred in enumerate(results["top_predictions"], 1):
-                bar_length = int(pred['confidence'] * 20)
-                bar = "█" * bar_length + "░" * (20 - bar_length)
-                output_text += f"{i}. **{pred['name']}** - {pred['confidence']:.1%}\n   `{bar}`\n\n"
-
-            output_text += f"""
----
-
-## 📊 Nutritional Information
-(per 100g serving)
-
-- **Calories:** {nutrition['calories']} kcal
-- **Protein:** {nutrition['protein']}g
-- **Carbohydrates:** {nutrition['carbs']}g
-- **Fat:** {nutrition['fat']}g
-- **Category:** {nutrition['category']}
-
----
-
-## 🖼️ Image Quality Analysis
-
-- **Quality Score:** {quality['quality_score']:.1f}/10
-- **Brightness:** {quality['brightness']:.0f}
-- **Saturation:** {quality['saturation']:.1f}
-- **Resolution:** {quality['width']}x{quality['height']}px
-
----
-
-## 🤖 Model Information
-
-- **Model:** {results['model_info']['model_type']}
-- **Dataset:** {results['model_info']['dataset']}
-- **Categories:** {results['model_info']['num_categories']}
-- **Device:** {results['model_info']['device']}
-"""
-
-            return image, output_text
-
-        except Exception as e:
-            logger.error(f"❌ Prediction error: {e}")
-            return None, f"❌ **Error:** {str(e)}\n\nPlease try another image."
-
-    # Health check funkcija za monitoring (API endpoint)
-    def health_check():
-        """Simple health check that returns OK status."""
-        return {"status": "healthy", "model_loaded": True}
-
-    # Kreiraj Gradio interfejs
-    with gr.Blocks(
-        title="AI Food Scanner",
-        theme=gr.themes.Soft()
-    ) as demo:
-
-        gr.Markdown("""
-# 🍽️ AI Food Scanner
-
-Upload an image of food to detect its type and get nutritional information.
-
-**Powered by EfficientNet-B0** trained on Food-101 dataset (101 food categories).
-        """)
-
-        with gr.Row():
-            with gr.Column(scale=1):
-                # Input: slika
-                input_image = gr.Image(
-                    label="📸 Upload Food Image",
-                    type="pil",
-                    sources=["upload", "clipboard"],
-                )
-
-                # Button za analizu
-                analyze_btn = gr.Button(
-                    "🔍 Analyze Food",
-                    variant="primary",
-                    size="lg"
-                )
-
-                # Primjeri (ako postoje)
-                gr.Examples(
-                    examples=[],  # Dodaj putanje do primjera ako imaš
-                    inputs=input_image,
-                    label="📋 Example Images"
-                )
-
-            with gr.Column(scale=1):
-                # Output: procesirana slika
-                output_image = gr.Image(
-                    label="🖼️ Processed Image",
-                    type="pil"
-                )
-
-                # Output: rezultati
-                output_text = gr.Markdown(
-                    label="📊 Analysis Results",
-                    value="*Results will appear here...*"
-                )
-
-        # Poveži button sa funkcijom
-        analyze_btn.click(
-            fn=predict_food,
-            inputs=input_image,
-            outputs=[output_image, output_text]
+    # Validate file type
+    if file.content_type not in ["image/jpeg", "image/jpg", "image/png", "image/webp"]:
+        raise HTTPException(
+            status_code=400,
+            detail="Invalid file type. Supported: JPEG, PNG, WebP"
         )
 
-        gr.Markdown("""
----
-
-## 📖 About This System
+    try:
+        # Read image
+        contents = await file.read()
+        image = Image.open(BytesIO(contents))
 
-This AI Food Scanner uses **EfficientNet-B0** model pretrained on the **Food-101** dataset.
+        # Predict
+        logger.info(f"🔍 Analyzing image: {file.filename}")
+        results = recognizer.predict(image, top_k=5)
 
-### Features:
-- ✅ 101 food categories recognition
-- ✅ ~85-90% accuracy
-- ✅ Nutritional information database
-- ✅ Image quality analysis
-- ✅ Optimized for CPU and GPU
-- ✅ Production-ready deployment
+        logger.info(f"✅ Prediction: {results['primary_prediction']['name']} ({results['primary_prediction']['confidence']:.2%})")
 
-### Technology Stack:
-- **Model:** EfficientNet-B0
-- **Framework:** PyTorch + Transformers
-- **Interface:** Gradio
-- **Deployment:** Hugging Face Spaces compatible
+        return JSONResponse(content=results)
 
-**Note:** Nutritional values are estimates per 100g serving.
-        """)
+    except Exception as e:
+        logger.error(f"❌ Error: {e}")
+        raise HTTPException(status_code=500, detail=f"Analysis failed: {str(e)}")
 
-    return demo
-
-
-# ==================== MAIN APPLICATION ====================
-def main():
-    """
-    Glavna funkcija - pokreće aplikaciju.
-    """
-    logger.info("=" * 80)
-    logger.info("🍽️ AI FOOD SCANNER - PRODUCTION READY SYSTEM")
-    logger.info("=" * 80)
-
-    # Selektuj device
-    device = select_device()
-
-    # Inicijalizuj model
-    logger.info("🚀 Initializing Food Recognition System...")
-    recognizer = FoodRecognizer(device)
-
-    # Kreiraj Gradio interfejs
-    logger.info("🎨 Creating Gradio Interface...")
-    demo = create_gradio_interface(recognizer)
+# ==================== MAIN ====================
+if __name__ == "__main__":
+    port = int(os.environ.get("PORT", 7860))
 
-    # Pokreni server
     logger.info("=" * 80)
-    logger.info("✅ System ready! Launching Gradio interface...")
+    logger.info("✅ API Ready!")
+    logger.info(f"📡 Server: http://0.0.0.0:{port}")
+    logger.info(f"📖 Docs: http://0.0.0.0:{port}/docs")
     logger.info("=" * 80)
 
-    # Launch sa konfiguracijom za Hugging Face Spaces
-    demo.launch(
-        server_name="0.0.0.0",
-        server_port=int(os.environ.get("PORT", 7860)),
-        share=False,
-        show_error=True,
-        auth=None  # No authentication required
+    uvicorn.run(
+        app,
+        host="0.0.0.0",
+        port=port,
+        log_level="info"
     )
-
-
-if __name__ == "__main__":
-    main()

requirements.txt

@@ -1,41 +1,26 @@
-# AI Food Scanner - Production Requirements
-# Optimized for Hugging Face Spaces (Free Tier: CPU/T4 GPU, 16-24GB RAM)
+# Food Recognition API - FastAPI Backend
+# Optimized for Hugging Face Spaces
 
-# ==================== Core Dependencies ====================
-# Python 3.9+ required
+# ==================== Core API Framework ====================
+fastapi>=0.104.0
+uvicorn[standard]>=0.24.0
+python-multipart>=0.0.6
 
-# Deep Learning Framework (upgraded for security CVE-2025-32434)
+# ==================== Deep Learning ====================
 torch>=2.6.0
 torchvision>=0.20.0
-
-# Hugging Face Transformers (for pretrained models)
 transformers>=4.35.0
 
-# ==================== UI Framework ====================
-# Gradio - Modern UI for ML demos
-gradio>=4.0.0
-
 # ==================== Image Processing ====================
-# PIL/Pillow - Image manipulation
 Pillow>=10.0.0
+numpy>=1.24.0,<2.0.0
 
-# NumPy - Numerical operations
-numpy>=1.21.0,<2.0.0
-
-# ==================== Optional Optimizations ====================
-# Accelerate - Faster model loading and inference
+# ==================== Optimizations ====================
 accelerate>=0.20.0
-
-# Safetensors - Faster model loading
 safetensors>=0.4.0
 
-# ==================== System Utilities ====================
-# Requests - HTTP library (backup dependencies)
-requests>=2.31.0
-
 # ==================== Notes ====================
-# - Total install size: ~2-3GB (PyTorch + models)
-# - Works on CPU and GPU (CUDA/MPS)
-# - Optimized for Hugging Face Spaces free tier
-# - No external API keys required - everything runs locally
-# - Models auto-download from Hugging Face Hub on first run
+# Model: nateraw/food (Food-101 pretrained)
+# Total size: ~2-3GB (PyTorch + model)
+# API endpoint: POST /api/analyze-food
+# CORS: Enabled for Next.js