app.py

import pandas as pd
import gradio as gr
import joblib

le=joblib.load('le_col.pkl')
min_max=joblib.load('MinMax.pkl')
lr=joblib.load('model.pkl')


le_col=['gender','Partner','Dependents','PhoneService','PaperlessBilling','MultipleLines','InternetService','OnlineSecurity','OnlineBackup','DeviceProtection','TechSupport','StreamingTV','StreamingMovies','Contract','PaymentMethod']
MinMax=['TotalCharges','MonthlyCharges','tenure']

def prediction_churn_model(gr,sc,pr,dp,t,ps,mp,Is,os,ob,dd,ts,st,sm,cr,pb,pm,mc,tc):
    try:
        input_data=pd.DataFrame({
            'gender':[gr],
            'SeniorCitizen':[sc],
            'Partner':[pr],
            'Dependents':[dp],
            'tenure':[t],
            'PhoneService':[ps],
            'MultipleLines':[mp],
            'InternetService':[Is],
            'OnlineSecurity':[os],
            'OnlineBackup':[ob],
            'DeviceProtection':[dd],
            'TechSupport':[ts],
            'StreamingTV':[st],
            'StreamingMovies':[sm],
            'Contract':[cr],
            'PaperlessBilling':[pb],
            'PaymentMethod':[pm],
            'MonthlyCharges':[mc],
            'TotalCharges':[tc]
        })
        for col in le_col:
            input_data[col]=le[col].transform(input_data[col])
        input_data[MinMax]=min_max.transform(input_data[MinMax])
        prediction=lr.predict(input_data)
        if prediction[0]==0:
            return 'No'
        else:
            return 'Yes'
    except Exception as e:
        return str(e)
gr.Interface(
    inputs=[
        gr.Dropdown(['Female', 'Male'],label='gender'),
        gr.Number(label='SeniorCitizen'),
        gr.Dropdown(['No','Yes'],label='Partner'),
        gr.Dropdown(['No','Yes'],label='Dependents'),
        gr.Number(label='tenure'),
        gr.Dropdown(['Yes','No'],label='PhoneService'),
        gr.Dropdown(['No','Yes'],label='MultipleLines'),
        gr.Dropdown(['Fiber optic','DSL','No'],label='InternetService'),
        gr.Dropdown(['No','Yes','No internet service'],label='OnlineSecurity'),
        gr.Dropdown(['No','Yes','No internet service'],label='OnlineBackup'),
        gr.Dropdown(['No','Yes','No internet service'],label='DeviceProtection'),
        gr.Dropdown(['No','Yes','No internet service'],label='TechSupport'),
        gr.Dropdown(['No','Yes','No internet service'],label='StreamingTV'),
        gr.Dropdown(['No','Yes','No internet service'],label='StreamingMovies'),
        gr.Dropdown(['Month-to-month','Two year','One year'],label='Contract'),
        gr.Dropdown(['Yes','No'],label='PaperlessBilling'),
        gr.Dropdown(['Electroniccheck', 'Mailedcheck', 'Banktransfer', 'Creditcard'],label='PaymentMethod'),
        gr.Number(label='MonthlyCharges'),
        gr.Number(label='TotalCharges')
    ],
    fn=prediction_churn_model,
    outputs=gr.Textbox(label='Prediction'),
    title='Prediction Telecoum Churn Model'
).launch()