models.py

import streamlit as st
import pandas as pd
import pickle
  

# Load the trained model
with open('model_xgb.pkl', 'rb') as file_1:
    model_inf_xgb = pickle.load(file_1)

# Function to run the Streamlit app
def run():
    # Set page title and sidebar image
    st.write('🔮 Customer Churn 🔮')

    # Introduction
    st.subheader("📊 Prediction of Customer Churn")
    st.write("Welcome to the Customer Churn Predictor app! This app predicts whether a customer will Churn on their subscription based on provided information.")

    # Input form
    st.markdown('## 📝 Input Data')
    with st.form('my_form'):
        # Input fields
        Age = st.number_input('💳 Age', min_value=0.0, max_value=65.0)
        st.markdown('**Gender:** Male And Female')
        Gender = st.selectbox('🚻 Gender', options=['Male', 'Female'])
        Tenure = st.number_input('💳 Tenure', min_value=0.0, max_value=60.0)
        Usage_Frequency = st.number_input('Usage Frequency', min_value=0.0, max_value=30.0)
        Support_Calls = st.number_input('Support Calls', min_value=0.0, max_value=10.0)
        Payment_Delay = st.number_input('Payment Delay', min_value=0.0, max_value=30.0)
        Subscription_Type = st.selectbox('Subscription Type', options=['Basic', 'Standard', 'Premium'])
        Contract_length = st.selectbox('Subscription Type', options=['Monthly', 'Annual', 'Quarterly'])
        Total_Spend = st.number_input('Total Spend', min_value=100.0, max_value=1000.0)

        submitted = st.form_submit_button('🔍 Let\'s Check!')

    # Create DataFrame from user input
    data = {
       'Age': Age,
       'Gender': Gender,
       'Tenure' : Tenure,
       'Usage Frequency' : Usage_Frequency,
       'Support Calls': Support_Calls,
       'Payment Delay' : Payment_Delay,
       'Subscription Type' : Subscription_Type,
       'Contract Length' : Contract_length,
       'Total Spend' : Total_Spend
    }

    df = pd.DataFrame([data])
    st.dataframe(df)

    # Make prediction
    if submitted:
        prediction = model_inf_xgb.predict(df)

        # Display prediction result
        if prediction[0] == 0:
            st.write('🟢 Not Churn')
        else:
            st.write('🔴 Churn')

if __name__== '__main__':
    run()