src/streamlit_app.py

import streamlit as st
import pandas as pd
import joblib

MODEL_PATH = 'src/credit_card_model.joblib'
SCALER_PATH = 'src/card_scaler.joblib' 

FEATURES = ['BALANCE', 'PURCHASES', 'CREDIT_LIMIT'] 

@st.cache_resource
def load_assets():
    try:
        model = joblib.load(MODEL_PATH)
        scaler = joblib.load(SCALER_PATH)
        return model, scaler
    except Exception as e:
        st.error(f"Error loading assets. Check if '{MODEL_PATH}' and '{SCALER_PATH}' are uploaded correctly. Error: {e}")
        return None, None

def predict_cluster(model, scaler, input_data):
    
    input_df = pd.DataFrame([input_data])
    
    scaled_data = scaler.transform(input_df[FEATURES])
    
    prediction = model.predict(scaled_data) 
    
    return prediction[0]

# --- Streamlit Interface ---
st.set_page_config(page_title="Credit Card Cluster Predictor", layout="wide")
st.title("💳 Credit Card Customer Segmentation")
st.markdown("Enter the customer's credit card usage details to predict their segment.")

model, scaler = load_assets()

if model is not None and scaler is not None:
    st.sidebar.header("Input Customer Data")

    balance = st.sidebar.slider("Current Balance ($):", min_value=0, max_value=20000, value=3000)
    purchases = st.sidebar.slider("Total Purchases ($):", min_value=0, max_value=15000, value=1000)
    credit_limit = st.sidebar.slider("Credit Limit ($):", min_value=1000, max_value=30000, value=5000)
    
    input_data = {
        'BALANCE': balance,
        'PURCHASES': purchases,
        'CREDIT_LIMIT': credit_limit
    }

    st.subheader("Customer Input Summary:")
    st.dataframe(pd.DataFrame([input_data]))

    if st.button("Predict Customer Segment"):
        with st.spinner('Predicting...'):
            cluster_id = predict_cluster(model, scaler, input_data)
            
            # Use simple descriptions (customize based on your 4 cluster names)
            cluster_descriptions = {
                0: "Cluster 0",
                1: "Cluster 1",
                2: "Cluster 2",
                3: "Cluster 3",
            }
            
            description = cluster_descriptions.get(cluster_id, f"🔍 Cluster ID **{cluster_id}** (Undefined Segment)")
            
            st.success("Prediction Successful!")
            st.markdown(f"## Predicted Segment: {description}")