commit 2

app.py

@@ -147,28 +147,37 @@ locations = {
     "Bachenbülach": 51,
     "Horgen": 295
 }
-def predict_apartment(rooms, area, town):
+def predict_apartment(rooms, area, proximity, town):
     bfs_number = locations[town]
     df = df_bfs_data[df_bfs_data['bfs_number']==bfs_number].copy()
     df.reset_index(inplace=True)
     df.loc[0, 'rooms'] = rooms
     df.loc[0, 'area'] = area
-    df.loc[0, 'proximity_to_public_transportation'] = 500  # Default value
+    df.loc[0, 'proximity_to_public_transportation'] = proximity  # Use user input instead of default
     
     if len(df) != 1:
         return -1
         
     features = ['rooms', 'area', 'pop', 'pop_dens', 'frg_pct', 'emp', 'tax_income', 'proximity_to_public_transportation']
-    X = df[features].values  # Convert to numpy array without feature names
+    X = df[features].values
     prediction = random_forest_model.predict(X)
     return np.round(prediction[0], 0)
 
 # Create the Gradio interface
 iface = gr.Interface(
     fn=predict_apartment,
-    inputs=["number", "number", gr.Dropdown(choices=locations.keys(), label="Town", type="value")],
-    outputs=[gr.Number()],
-    examples=[[4.5, 120, "Dietlikon"], [3.5, 60, "Winterthur"]]
+    inputs=[
+        gr.Number(label="Number of Rooms"),
+        gr.Number(label="Area (m²)"),
+        gr.Slider(minimum=0, maximum=2000, value=500, step=50, 
+                 label="Distance to Public Transportation (meters)"),
+        gr.Dropdown(choices=locations.keys(), label="Town", type="value")
+    ],
+    outputs=[gr.Number(label="Predicted Price (CHF)")],
+    examples=[
+        [4.5, 120, 500, "Dietlikon"], 
+        [3.5, 60, 250, "Winterthur"]
+    ],
+    description="Predict apartment prices in Zürich based on rooms, area, proximity to public transportation, and location."
 )
-
 iface.launch()