telco_churn / app.py
op1009
add app files
014959f
import gradio as gr
import pickle
import pandas as pd
import numpy as np
# load the saved model and DictVectorizer
with open('dict_vectorizer.bin', 'rb') as f_in:
loaded_dict_vec = pickle.load(f_in)
with open('model_lr.bin', 'rb') as f_in:
loaded_model = pickle.load(f_in)
# Function to make predictions
def predict_churn(gender, seniorcitizen, partner, dependents, phoneservice,
multiplelines, internetservice, onlinesecurity, onlinebackup,
deviceprotection, techsupport, streamingtv, streamingmovies,
contract, paperlessbilling, paymentmethod, tenure,
monthlycharges, totalcharges):
customer = {
'gender': gender,
'seniorcitizen': seniorcitizen,
'partner': partner,
'dependents': dependents,
'phoneservice': phoneservice,
'multiplelines': multiplelines,
'internetservice': internetservice,
'onlinesecurity': onlinesecurity,
'onlinebackup': onlinebackup,
'deviceprotection': deviceprotection,
'techsupport': techsupport,
'streamingtv': streamingtv,
'streamingmovies': streamingmovies,
'contract': contract,
'paperlessbilling': paperlessbilling,
'paymentmethod': paymentmethod,
'tenure': tenure,
'monthlycharges': monthlycharges,
'totalcharges': totalcharges
}
labels = ['No Churn', 'Churn']
X = loaded_dict_vec.transform([customer])
y_pred = loaded_model.predict_proba(X)[0,:]
churn_probability = y_pred
return {labels[i]: float(y_pred[i]) for i in range(2)}
with gr.Blocks() as demo:
gr.Markdown("<h1 style='text-align: center;'>Telecom Customer Churn Prediction</h1>")
gr.Markdown("<h3 style='text-align: center;'>Predict the probability of a customer churning based on their demographics and service usage.</h3>")
with gr.Row():
gender = gr.components.Dropdown(['male', 'female'], value='male', label="Gender")
senior_citizen = gr.components.Radio([0, 1], value=0, label="Senior Citizen")
partner = gr.components.Radio(['yes', 'no'], value='yes', label="Partner")
dependents = gr.components.Radio(['yes', 'no'], value='yes', label="Dependents")
with gr.Row():
phone_service = gr.components.Radio(['yes', 'no'], value='yes', label="Phone Service")
multiple_lines = gr.components.Dropdown(['no_phone_service', 'no', 'yes'], label="Multiple Lines")
internet_service = gr.components.Dropdown(['dsl', 'fiber_optic', 'no'], label="Internet Service")
online_security = gr.components.Dropdown(['no', 'yes', 'no_internet_service'], label="Online Security")
online_backup = gr.components.Dropdown(['yes', 'no', 'no_internet_service'], label="Online Backup")
device_protection = gr.components.Dropdown(['no', 'yes', 'no_internet_service'], label="Device Protection")
tech_support = gr.components.Dropdown(['no', 'yes', 'no_internet_service'], label="Tech Support")
with gr.Row():
streaming_tv = gr.Dropdown(label="Streaming TV", choices=['no', 'yes', 'no_internet_service'], value="no")
streaming_movies = gr.Dropdown(label="Streaming Movies", choices=['no', 'yes', 'no_internet_service'], value="no")
with gr.Row():
contract = gr.components.Dropdown(['month-to-month', 'one_year', 'two_year'], value='month-to-month', label="Contract")
paperless_billing = gr.components.Radio(['yes', 'no'], value='no', label="Paperless Billing")
payment_method = gr.components.Dropdown(['electronic_check', 'mailed_check', 'bank_transfer_(automatic)', 'credit_card_(automatic)'], value='electronic_check', label="Payment Method")
tenure = gr.components.Number(label="Tenure (Months)", minimum=0, maximum=1000)
monthly_charges = gr.components.Number(label="Monthly Charges (₹)", minimum=0, maximum=10000)
total_charges = gr.components.Number(label="Total Charges (₹)", minimum=0, maximum=1000000)
submit_btn = gr.Button("Submit")
output_label = gr.Label(label="Churn Probability")
submit_btn.click(predict_churn, inputs=[gender, senior_citizen, partner, dependents,
phone_service, multiple_lines,
internet_service, online_security, online_backup, device_protection, tech_support,
streaming_tv, streaming_movies, contract, paperless_billing, payment_method, tenure,
monthly_charges, total_charges
], outputs=output_label)
demo.launch()