Update app.py

app.py

@@ -11,7 +11,6 @@ import base64
 
 app = FastAPI(title="Waste Classification API")
 
-# Add CORS middleware
 app.add_middleware(
     CORSMiddleware,
     allow_origins=["*"],
@@ -20,12 +19,11 @@ app.add_middleware(
     allow_headers=["*"],
 )
 
-# Load the model (you'll need to place your model file in the api directory)
 try:
     model = load_model("predictWaste12.h5")
 except:
     model = None
-    print("Model not found. Please place predictWaste12.h5 in the api directory.")
+    print("Model not found.")
 
 output_class = [
     "baterai", "biologis", "kaca-coklat", "kardus", "pakaian", 
@@ -40,21 +38,12 @@ output_class_english = [
 ]
 
 def preprocess_image(img_bytes):
-    """Preprocess image for prediction"""
-    # Convert bytes to PIL Image
     img = Image.open(io.BytesIO(img_bytes))
-    
-    # Convert to RGB if necessary
     if img.mode != 'RGB':
         img = img.convert('RGB')
-    
-    # Resize to model input size
     img = img.resize((224, 224))
-    
-    # Convert to array and normalize
     img_array = np.array(img) / 255.0
     img_array = np.expand_dims(img_array, axis=0)
-    
     return img_array
 
 @app.get("/")
@@ -65,28 +54,20 @@ async def root():
 async def predict_waste(file: UploadFile = File(...)):
     if not model:
         raise HTTPException(status_code=500, detail="Model tidak tersedia")
-    
     try:
-        # Read image file
         img_bytes = await file.read()
-        
-        # Preprocess image
         processed_image = preprocess_image(img_bytes)
-        
-        # Make prediction
         predicted_array = model.predict(processed_image)
         predicted_index = np.argmax(predicted_array)
         predicted_class = output_class[predicted_index]
         predicted_class_en = output_class_english[predicted_index]
         predicted_accuracy = round(np.max(predicted_array) * 100, 2)
-        
         return {
             "prediction": predicted_class,
             "prediction_en": predicted_class_en,
             "accuracy": predicted_accuracy,
             "confidence_scores": predicted_array[0].tolist()
         }
-        
     except Exception as e:
         raise HTTPException(status_code=500, detail=f"Error processing image: {str(e)}")