segmentation.py

import streamlit as st
import pandas as pd
import joblib

# Load trained model and scaler
kmeans = joblib.load('kmeans_model.pkl')
scaler = joblib.load('scaler.pkl')

st.title('Customer Segmentation App')
st.write('Enter customer details to predict the customer segment.')

# User inputs
age = st.number_input('Age', min_value=18, max_value=100, value=35)
income = st.number_input('Income', min_value=0, max_value=200000, value=50000)
total_spending = st.number_input(
    'Total Spending (sum of purchases)', min_value=0, max_value=50000, value=1000
)
num_web_purchases = st.number_input(
    'Number of Web Purchases', min_value=0, max_value=100, value=10
)
num_store_purchases = st.number_input(
    'Number of Store Purchases', min_value=0, max_value=100, value=10
)
num_web_visits = st.number_input(
    'Number of Web Visits per Month', min_value=0, max_value=50, value=3
)
recency = st.number_input(
    'Recency (days since last purchase)', min_value=0, max_value=365, value=30
)

# Input dataframe (FEATURE NAMES MUST MATCH TRAINING)
input_data = pd.DataFrame({
    'Age': [age],
    'Income': [income],
    'Total_Spending': [total_spending],
    'NumWebPurchases': [num_web_purchases],
    'NumStorePurchases': [num_store_purchases],
    'NumWebVisitsMonth': [num_web_visits],
    'Recency': [recency]
})

# Scale input
input_scaled = scaler.transform(input_data)

# Prediction
if st.button('Predict Segment'):
    cluster = kmeans.predict(input_scaled)
    st.success(f'Predicted Segment: Cluster {cluster[0]}')