Update app.py

app.py

@@ -9,18 +9,23 @@ key=os.environ["HF_KEY"]
 
 
 def load_model():
-    pipe=pipeline(task="fill-mask",model="BounharAbdelaziz/XLM-RoBERTa-Morocco",token=key,device=0)
+    print("[INFO] Loading model... This may take a minute on Spaces")
+    pipe = pipeline(
+        task="fill-mask",
+        model="atlasia/XLM-RoBERTa-Morocco",
+        token=key,
+        device=0,
+        torch_dtype=torch.float16  # Use half precision
+    )
+    print("[INFO] Model loaded successfully!")
     return pipe
 
 print("[INFO] load model ...")
 pipe=load_model()
 print("[INFO] model loaded")
 
-# def predict(text):
-#     predictions=pipe(text)
-#     return predictions[0]["sequence"],predictions
-
 @spaces.GPU
+@gr.cache(persist=True)  # Add persistent caching
 def predict(text):
     outputs = pipe(text)
     scores= [x["score"] for x in outputs]
@@ -43,24 +48,28 @@ with gr.Blocks() as demo:
                 clear_btn = gr.Button("Clear")
                 submit_btn = gr.Button("Submit", variant="primary")
             
-            # Examples section
+            # Examples section with caching
             gr.Examples(
                 examples=["العاصمة د <mask> هي الرباط","المغرب <mask> زوين","انا سميتي مريم، و كنسكن ف<mask> العاصمة دفلسطين"],
-                inputs=input_text
+                inputs=input_text,
+                cache_examples=True,
+                preprocess=True  # Precompute examples
             )
         
         with gr.Column():
             # Output probabilities
             output_labels = gr.Label(
                 label="Prediction Results",
-                show_label=False
+                show_label=False,
+                num_top_classes=5  # Limit to top 5 predictions
             )
 
     # Button actions
     submit_btn.click(
         predict,
         inputs=input_text,
-        outputs=output_labels
+        outputs=output_labels,
+        show_progress=True  # Show a progress indicator
     )
     
     clear_btn.click(
@@ -68,5 +77,6 @@ with gr.Blocks() as demo:
         outputs=input_text
     )
 
-# Launch the app
-demo.launch()
+# Launch the app with queue
+demo.queue(concurrency_count=3)  # Allow 3 concurrent predictions
+demo.launch(show_api=False)  # Disable API tab if not needed