app.py

import pandas as pd,numpy as np
import pickle
import streamlit as st

# st.title("Mobile Data User Behaviour")

# st.title("User Behavior Using Mobile Prediction")
# model_2 = pickle.load(open(r"\rfc.pkl","rb")) #pickle file path

# if st.button("Submit"):
#     result = model_2.predict([[]])
#     st.write(f"The predicted price of the rental house is {result}")

# import joblib
# import streamlit as st

# st.title("User Behavior Using Mobile Prediction")

# # Inputs
# Device_Model = st.selectbox("Enter Device Model Type", ['Xiaomi Mi 11', 'iPhone 12', 'Google Pixel 5', 'OnePlus 9', 'Samsung Galaxy S21'])
# age = st.number_input("Enter age", min_value=0, max_value=100, step=1, format="%d")
# gender = st.radio("Enter gender", ['Male', 'Female'])
# Operating_System = st.selectbox("Enter Operating System Type", ['Android', 'iOS'])
# App_Usage_Time = st.number_input("Enter the App Usage Time")
# Screen_On_Time = st.number_input("Enter the Screen On Time")
# 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")

# # Load the model
# model_path = "./rfc.pkl"  # Adjust if in a subdirectory
# try:
#     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 FileNotFoundError:
#     st.error(f"Model file not found at: {model_path}")
# except Exception as e:
#     st.error(f"An error occurred: {e}")


import joblib
import streamlit as st

# Set page configuration
st.set_page_config(page_title="User Behavior Prediction", page_icon="📱", layout="wide")

# Title and Description
st.title("📱 User Behavior Prediction Using Mobile Data")
st.markdown("""
Welcome to the **User Behavior Prediction App**!  
This tool predicts user behavior based on mobile usage data, powered by a machine learning model.  
Fill in the details below and click **Submit** to see the results.
""")

# Inputs in columns for better alignment
col1, col2 = st.columns(2)

with col1:
    Device_Model = st.selectbox(
        "📱 Device Model Type",
        ['Xiaomi Mi 11', 'iPhone 12', 'Google Pixel 5', 'OnePlus 9', 'Samsung Galaxy S21'],
        help="Select the type of mobile device being used."
    )
    Operating_System = st.selectbox(
        "⚙️ Operating System Type",
        ['Android', 'iOS'],
        help="Select the mobile's operating system."
    )
    gender = st.radio(
        "👤 Gender",
        ['Male', 'Female'],
        help="Select the gender of the user."
    )
    age = st.number_input(
        "🎂 Age",
        min_value=0, max_value=100, step=1, format="%d",
        help="Enter the user's age."
    )

with col2:
    App_Usage_Time = st.number_input(
        "⏱️ App Usage Time (in hours)",
        min_value=0.0, step=0.1,
        help="Enter the total app usage time (e.g., 3.5 hours)."
    )
    Screen_On_Time = st.number_input(
        "🔆 Screen On Time (in hours)",
        min_value=0.0, step=0.1,
        help="Enter the total screen on time (e.g., 5.2 hours)."
    )
    Battery_Drain = st.number_input(
        "🔋 Battery Drain (in percentage)",
        min_value=0.0, step=0.1,
        help="Enter the battery drain percentage (e.g., 15.5%)."
    )
    Number_of_Apps_Installed = st.number_input(
        "📱 Number of Apps Installed",
        min_value=0, step=1,
        help="Enter the total number of apps installed on the device."
    )
    Data_Usage = st.number_input(
        "📶 Data Usage (in GB)",
        min_value=0.0, step=0.1,
        help="Enter the total data usage (e.g., 1.5 GB)."
    )

# Load the model
model_path = "./rfc.pkl"  # Adjust if in a subdirectory
try:
    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]

    # Predict and display result
    if st.button("🚀 Submit"):
        with st.spinner("Running prediction..."):
            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.success("🎉 Prediction Completed!")
        st.markdown(f"**📊 Predicted User Behavior:** `{result[0]}`")

except FileNotFoundError:
    st.error(f"🚨 Model file not found at: `{model_path}`. Please upload the model.")
except Exception as e:
    st.error(f"🚨 An error occurred: {e}")



    

# age = st.number_input("Enter age",min_value=0,max_value=1000,step=1,format="%d" )
# gender = st.radio("Enter gender",['Male','Female'])
# chestpain = st.selectbox("chestpain",['non-anginal_pain', 'typical_angina', 'atypical_angina','asymptomatic'])
# restingBP =  st.number_input("Enter BP",min_value=0,max_value=1000,step=1,format="%d")
# serum_cholesterol = st.number_input("Enter serum_cholesterol",min_value=0,max_value=10000,step=1,format="%d")
# fasting_blood_sugar = st.radio("Enter fasting_blood_sugar",['yes','no'])
# restingrelectro= st.selectbox("Enter resting_electro",['ST-T_wave_abnormality', 'normal', 'left_ventricular_hypertrophy'])
# maxheartrate = st.number_input("Enter max_heart_rate",min_value=0,max_value=1000,step=1,format="%d")
# exerciseangia =  st.radio("Enter exercise_angia",['yes','no'])
# oldpeak= st.number_input("Enter oldpeak")
# slope = st.selectbox("Enter slope",['downsloping', 'upsloping', 'flat'])
# noofmajorvessels = st.selectbox("enter number of major vessels",['Three', 'One', 'Zero', 'Two'])