import streamlit as st
import requests
import pandas as pd

# Backend URL
BACKEND_URL = "https://omm7-boston-backend.hf.space"

# Check backend health
try:
    health_resp = requests.get(f"{BACKEND_URL}/health", timeout=5)
    if health_resp.status_code == 200 and health_resp.json().get("status") == "ok":
        st.toast("✅ Connected to backend", icon="✅")
    else:
        st.toast("⚠️ Backend connection failed", icon="⚠️")
except Exception:
    st.toast("❌ Backend not reachable", icon="❌")

st.title('Boston Housing Price Predictor 🏠')
st.write('Enter the details of the area to predict the median home value.')

# Input fields
crim = st.number_input('Per capita crime rate (CRIM)', value=0.0)
zn = st.number_input('Proportion of residential land zoned for lots over 25,000 sq.ft. (ZN)', value=0.0)
indus = st.number_input('Proportion of non-retail business acres per town (INDUS)', value=0.0)
chas = st.selectbox('Tract bounds Charles River? (CHAS)', options=[0, 1], format_func=lambda x: 'Yes' if x == 1 else 'No')
nox = st.number_input('Nitric oxides concentration (NOX)', value=0.0)
rm = st.number_input('Average number of rooms per dwelling (RM)', value=0.0)
age = st.number_input('Proportion of owner-occupied units built prior to 1940 (AGE)', value=0.0)
dis = st.number_input('Weighted distances to five Boston employment centers (DIS)', value=0.0)
rad = st.number_input('Index of accessibility to radial highways (RAD)', value=0.0)
tax = st.number_input('Full-value property-tax rate per $10,000 (TAX)', value=0.0)
ptratio = st.number_input('Pupil-teacher ratio by town (PTRATIO)', value=0.0)
lstat = st.number_input('% lower status of the population (LSTAT)', value=0.0)

input_data = {
    'CRIM': crim,
    'ZN': zn,
    'INDUS': indus,
    'CHAS': chas,
    'NOX': nox,
    'RM': rm,
    'AGE': age,
    'DIS': dis,
    'RAD': rad,
    'TAX': tax,
    'PTRATIO': ptratio,
    'LSTAT': lstat
}

if st.button('Predict Median Home Value'):
    status_placeholder = st.empty()
    status_placeholder.info("📡 Sending request to backend...")

    try:
        response = requests.post(f"{BACKEND_URL}/predict", json=input_data)
        if response.status_code == 200:
            data = response.json()

            # Show step-by-step status
            for step in data.get("steps", []):
                status_placeholder.info(step)

            prediction = data['prediction']
            st.success(f'The predicted median home value is: ${prediction:.2f} (in thousands)')
            st.markdown(f'**Predicted Value**: ${prediction * 1000:,.2f}')
        else:
            st.error(f"Error from API: {response.text}")
    except requests.exceptions.ConnectionError:
        st.error("❌ Connection error. Please ensure the backend is running.")

st.markdown("---")
st.write("This application uses a machine learning model to predict the median value of homes in Boston suburbs.")