Update app.py

app.py

@@ -22,8 +22,7 @@
 #     st.write(f"The predicted price of the rental house is {result}")
 
 
-import pandas as pd
-import numpy as np
+import os
 import joblib
 import streamlit as st
 
@@ -39,19 +38,29 @@ Battery_Drain = st.number_input("Enter the Battery Drain")
 Number_of_Apps_Installed = st.number_input("Enter the Number of Apps Installed")
 Data_Usage = st.number_input("Enter the Data Usage")
 
-try:
-    model_2 = joblib.load(r"Mobile_data_user_behaviour/rfc.pkl")  # Adjust path as needed
-except FileNotFoundError:
-    st.error("Model file 'rfc.pkl' not found. Please check the file path.")
-
-if st.button("Submit"):
+# Path Debugging
+model_path = r"Mobile_data_user_behaviour/rfc.pkl"
+if not os.path.exists(model_path):
+    st.error(f"Model file not found at: {model_path}")
+else:
     try:
-        # Encode categorical features if necessary
-        # Assuming you preprocess the data before prediction
-        result = model_2.predict([[Device_Model, Operating_System, App_Usage_Time, Screen_On_Time, Battery_Drain, Number_of_Apps_Installed, Data_Usage, age, gender]])
-        st.write(f"The predicted price of the rental house is {result}")
+        model_2 = joblib.load(model_path)
+
+        # Encode categorical variables
+        mapping_device = {'Xiaomi Mi 11': 0, 'iPhone 12': 1, 'Google Pixel 5': 2, 'OnePlus 9': 3, 'Samsung Galaxy S21': 4}
+        mapping_os = {'Android': 0, 'iOS': 1}
+        mapping_gender = {'Male': 0, 'Female': 1}
+
+        device_model_encoded = mapping_device[Device_Model]
+        operating_system_encoded = mapping_os[Operating_System]
+        gender_encoded = mapping_gender[gender]
+
+        if st.button("Submit"):
+            result = model_2.predict([[device_model_encoded, operating_system_encoded, App_Usage_Time, Screen_On_Time, Battery_Drain, Number_of_Apps_Installed, Data_Usage, age, gender_encoded]])
+            st.write(f"The predicted behavior is: {result}")
     except Exception as e:
-        st.error(f"An error occurred during prediction: {e}")
+        st.error(f"Failed to load the model or predict: {e}")
+