imastanaya08/Chustomer_Churn_Prediction

1import streamlit as st2import tensorflow as tf3from tensorflow import keras4from tensorflow.keras.models import load_model5from keras import layers6import numpy as np7import pandas as pd8import pickle9import json10 11#Load All File12with open('final_pipeline.pkl', 'rb') as file_1:13  model_pipeline = pickle.load(file_1)14 15model_ann = load_model('churn_model.h5')16 17def run() :18     19    # Membuat Form20    with st.form(key='form_parameters'):21        user_id = st.text_input('user_id', value='')22        age = st.number_input('Age', min_value=16, max_value=90, value=25, step=1, help='Usia Pemain')23        gender = st.selectbox('Gender', ('Male', 'Female'), index=1)24        region_category = st.selectbox('Region', ('City', 'Velage','Town'), index=1)25        membership_category = st.selectbox('Membership', ('No Membership', 'Basic Membership','Silver Membership','Gold Membership'), index=1)26        joining_date = st.selectbox('Joining', ('2015','2016','2017','2018','2019','2020', '2022','2023'), index=1)27        joined_through_referral = st.selectbox('joined_through_referral', ('Yes', 'No'), index=1)28        st.markdown('---')29 30        preferred_offer_types = st.selectbox('preferred_offer_types', ('Without Offers', 'Credit/Debit Card Offers','Gift Vouchers/Coupons'), index=1)31        medium_of_operation = st.selectbox('medium_of_operation', ('Desktop', 'Smartphone','Both'), index=1)32        internet_option = st.selectbox('internet_option', ('Wi-Fi', 'Fiber_Optic','Mobile_Data'), index=1)33        last_visit_time = st.number_input('last_visit_time', min_value=0, max_value=100, value=50)34        days_since_last_login = st.slider('days_since_last_login', -900,17, 100)35        avg_time_spent = st.slider('avg_time_spent', 0,100,500)36 37        avg_transaction_value = st.slider('avg_transaction_value', 0,10000,20000)38        avg_frequency_login_days = st.slider('avg_frequency_login_days', 0,11,200)39        points_in_wallet = st.slider('points_in_wallet', 0,590,2000)40        used_special_discount = st.selectbox('used_special_discount', ('Yes', 'No'), index=1)41        offer_application_preference = st.selectbox('offer_application_preference', ('Yes', 'No'), index=1)42        past_complaint = st.selectbox('past_complaint', ('Yes', 'No'), index=1)43 44        complaint_status = st.selectbox('complaint_status', ('Not Applicable', 'Solved', 'Solved in Follow', 'Unsolved','No Information Available'), index=1)45        feedback = st.selectbox('feedback', ('Not Poor Website', 'Poor Customer Service', 'Too many ads', 'Poor Product Quality','Reasonable Price','User Friendly Website'), index=1)46        churn_risk_score = st.selectbox('churn_risk_score', ('0', '1'), index=1)47 48        submitted = st.form_submit_button('Predict')49 50    data_inf = {51          'user_id ': '8f420209e7d129f8',52          'age': 29,53          'gender': 'F',54          'region_category': 'City',55          'membership_category': 'Basic Membership',56          'joining_date ': 2017-2-11,57          'joined_through_referral': 'Yes',58          'preferred_offer_types ': 'Yes',59          'medium_of_operation ': 'Desktop',60          'internet_option ': 'Fiber_Optic',61          'last_visit_time ' : 17,62          'days_since_last_login' : 17,63          'avg_time_spent': 338.150000,64          'avg_transaction_value': 15678.14,65          'avg_frequency_login_days': 11.0,66          'points_in_wallet': 590.22,67          'used_special_discount': 'No',68          'offer_application_preference': 'Yes',69          'past_complaint': 'Yes',70          'complaint_status': 'No Information Available',71          'feedback': 'Poor Customer Service',72          'churn_risk_score': '1'73    }74    data_inf = pd.DataFrame([data_inf])75    st.dataframe(data_inf)76 77    if submitted:78        # Transform Inference-Set79 80        data_inf_transform = model_pipeline.transform(data_inf)81        data_inf_transform82 83        # Predict using Neural Network84        y_pred_inf = model_ann.predict(data_inf_transform)85        y_pred_inf = np.where(y_pred_inf>= 0.5,1,0)86        if y_pred_inf == 1:87 88          st.write('### Hasil Prediksi apakah pelanggan Churn : Yes ')89 90        else:91          st.write('### Hasil Prediksi apakah pelanggan Churn : No ')92