Update app.py

app.py

@@ -201,6 +201,12 @@ def load_image(img_path, preprocess=True, height=320, width=320):
         x = np.expand_dims(x, axis=0)
     return x
 
+def rename_layers(model):
+    for layer in model.layers:
+        if '/' in layer.name:
+            layer._name = layer.name.replace('/', '_')
+    return model
+
 def grad_cam(input_model, img_array, cls, layer_name):
     grad_model = tf.keras.models.Model(
         [input_model.inputs],
@@ -241,7 +247,17 @@ def compute_gradcam(model_gradcam, img_path, layer_name='bn'):
     # model_gradcam.compile(optimizer=tf.keras.optimizers.Adam(learning_rate=0.0001),
     #           loss='sparse_categorical_crossentropy')
     # model.load_weights('./pretrained_model.h5')
-    model_gradcam = keras.models.load_model('./model_renamed.h5')
+    # Load the original model
+    original_model = keras.models.load_model('./gradcam_model.h5')
+    
+    # Rename the layers
+    modified_model = rename_layers(original_model)
+    
+    # Save the modified model
+    modified_model.save('./modified_gradcam_model.h5')
+    
+    # Now use this modified model in your application
+    model_gradcam = keras.models.load_model('./modified_gradcam_model.h5')
     
     preprocessed_input = load_image(img_path)
     predictions = model_gradcam.predict(preprocessed_input)
@@ -493,7 +509,7 @@ if uploaded_file is not None:
     with col3:
         if st.button('Generate Grad-CAM'):
             st.write("Loading model...")
-            model_gradcam = keras.models.load_model('./gradcam_model.h5')
+            model_gradcam = keras.models.load_model('./modified_gradcam_model.h5')
             # Compute and show Grad-CAM
             st.write("Generating Grad-CAM visualizations")
             try: