Update src/streamlit_app.py

src/streamlit_app.py

@@ -11,19 +11,19 @@ IMG_SIZE = 300
 CLASS_NAMES = ['cup', 'fork', 'glass', 'knife', 'plate', 'spoon']
 
 # -----------------------------------
-# MODEL LOADING
+# MODEL LOADING (KERAS 3)
 # -----------------------------------
 @st.cache_resource
 def load_model_file():
     model_path = os.path.join(os.path.dirname(__file__), "kitchen1.keras")
-    # TF 2.14'te de bu yöntem çalışır
-    return tf.keras.models.load_model(model_path, compile=False)
+    try:
+        # Keras 3'te model yükleme
+        return tf.keras.models.load_model(model_path, compile=False)
+    except Exception as e:
+        st.error(f"Model yüklenemedi: {e}")
+        return None
 
-try:
-    model = load_model_file()
-except Exception as e:
-    st.error(f"Model yüklenirken hata: {e}")
-    st.stop()
+model = load_model_file()
 
 # -----------------------------------
 # IMAGE PREPROCESS
@@ -31,12 +31,19 @@ except Exception as e:
 def process_image(img):
     img = img.resize((IMG_SIZE, IMG_SIZE))
     img = np.array(img)
+    
+    # Boyut kontrolleri
     if img.ndim == 2:
         img = np.stack([img]*3, axis=-1)
     elif img.shape[-1] == 4:
         img = img[..., :3]
     
+    # Batch boyutu ekle: (1, 300, 300, 3)
     img = np.expand_dims(img, 0)
+    
+    # EfficientNet B3 Preprocess
+    # Keras 3'te bu fonksiyon bazen farklı path'te olabilir,
+    # ama tf.keras.applications genelde güvenlidir.
     img = tf.keras.applications.efficientnet.preprocess_input(img)
     return img
 
@@ -48,9 +55,10 @@ st.write("Upload an image to classify kitchen items using EfficientNetB3.")
 
 uploaded = st.file_uploader("Upload an image", type=["jpg", "jpeg", "png"])
 
-if uploaded:
+if uploaded and model is not None:
     try:
         img = Image.open(uploaded).convert("RGB")
+        
         col1, col2 = st.columns(2)
         
         with col1:
@@ -58,6 +66,7 @@ if uploaded:
             
         with col2:
             st.write("Classifying...")
+            
             image_tensor = process_image(img)
             preds = model.predict(image_tensor)
             
@@ -68,5 +77,6 @@ if uploaded:
             st.success(f"Prediction: **{predicted_class.upper()}**")
             st.metric("Confidence", f"{confidence * 100:.2f}%")
             st.progress(int(confidence * 100))
+            
     except Exception as e:
-        st.error(f"Hata: {e}")
+        st.error(f"Hata oluştu: {e}")