Update app.py

app.py

@@ -3,17 +3,17 @@ import pickle
 import numpy as np
 from sklearn.preprocessing import StandardScaler
 
-# Load the model at startup
+# 1) Load your model once at startup
 with open("rf_model.pkl", "rb") as f:
     MODEL = pickle.load(f)
 
-# Prediction function
+# 2) A simple predict function
 def predict(f1, f2, f3, f4, f5):
     X = np.array([[f1, f2, f3, f4, f5]])
     X_scaled = StandardScaler().fit_transform(X)
     return int(MODEL.predict(X_scaled)[0])
 
-# Build the interface
+# 3) Build your Gradio interface
 demo = gr.Interface(
     fn=predict,
     inputs=[
@@ -24,11 +24,15 @@ demo = gr.Interface(
         gr.Slider(0, 10, step=1, value=4, label="Feature 5"),
     ],
     outputs=gr.Label(label="Prediction"),
-    title="TaskMaster Job Scheduler",
-    description="RandomForest inference on 5 synthetic features",
+    title="TaskMaster Job Scheduler",
+    description="Enter five feature values to get a RandomForest prediction."
 )
 
 if __name__ == "__main__":
-    # Bind to 0.0.0.0 so Spaces can route traffic
-    # block=True keeps the Python process alive (no immediate exit)
-    demo.launch(server_name="0.0.0.0", server_port=7860, block=True)
+    # This tells Gradio to bind to all interfaces, on port 7860,
+    # and to BLOCK the Python thread (so the container stays up).
+    demo.launch(
+        server_name="0.0.0.0",
+        server_port=7860,
+        prevent_thread_lock=False
+    )