app.py

# Streamlit app supporting RF and XGB models
import streamlit as st
import pandas as pd, os, joblib

st.title('Tourism Package Prediction - Demo (RF & XGB)')

MODEL_RF = '/data/output/best_model.joblib'
MODEL_XGB = '/data/output/best_xgb.joblib'

models = {}
if os.path.exists(MODEL_RF):
    models['RandomForest'] = joblib.load(MODEL_RF)
if os.path.exists(MODEL_XGB):
    models['XGBoost'] = joblib.load(MODEL_XGB)

if not models:
    st.error('No model artifacts found. Place best_model.joblib or best_xgb.joblib in /mnt/data/output/')
    st.stop()

choice = st.selectbox('Choose model', list(models.keys()))
model = models[choice]

st.write('Selected model:', choice)
st.write('Upload a CSV (features only) to predict:')
uploaded = st.file_uploader('Input CSV', type=['csv'])
if uploaded:
    df = pd.read_csv(uploaded)
    st.write('Preview:', df.head())
    preds = model.predict(df)
    st.write('Predictions:', preds)