app.py

import streamlit as st
import pandas as pd
import plotly.express as px
from Model.heat_model import train_and_predict

st.set_page_config(page_title="Urban Heat Island AI", layout="wide")

st.title("🌆🔥 Urban Heat Island AI")
st.write("AI-powered heat risk prediction for sustainable cities")

uploaded_file = st.file_uploader("Upload City Heat Dataset (CSV)", type=["csv"])

if uploaded_file:
    df = pd.read_csv(uploaded_file)

    st.subheader("📊 Raw Data")
    st.dataframe(df)

    df = train_and_predict(df)

    st.subheader("🔥 Heat Risk Prediction")

    fig = px.scatter_mapbox(
        df,
        lat="latitude",
        lon="longitude",
        color="heat_risk",
        size="heat_risk",
        color_continuous_scale="YlOrRd",
        zoom=10,
        mapbox_style="open-street-map",
        title="Urban Heat Risk Map"
    )

    st.plotly_chart(fig, use_container_width=True)

    st.subheader("🌱 Cooling Simulation")

    trees_added = st.slider("Increase Green Cover (%)", 0, 30, 10)

    df['simulated_risk'] = df['heat_risk'] - (trees_added * 0.8)
    df['simulated_risk'] = df['simulated_risk'].clip(lower=0)

    st.write("Heat risk after adding green cover:")

    fig2 = px.scatter_mapbox(
        df,
        lat="latitude",
        lon="longitude",
        color="simulated_risk",
        size="simulated_risk",
        color_continuous_scale="Greens",
        zoom=10,
        mapbox_style="open-street-map",
        title="Post-Intervention Heat Risk"
    )

    st.plotly_chart(fig2, use_container_width=True)

else:
    st.info("👆 Upload a CSV file to start the demo")