import pandas as pd
import streamlit as st
import pickle
from PIL import Image

# Set page configuration
st.set_page_config(
    page_title="Electric Bill Predictor",
    page_icon="⚡",
    layout="centered",
    initial_sidebar_state="expanded"
)

# Custom CSS
st.markdown("""
    <style>
        .title {
            color: #2c3e50;
            text-align: center;
            margin-bottom: 30px;
        }
        .input-section {
            background-color: #f8f9fa;
            padding: 20px;
            border-radius: 10px;
            margin-bottom: 20px;
            border-left: 5px solid #27ae60;
        }
        .result-box {
            background-color: #e8f5e9;
            padding: 25px;
            border-radius: 10px;
            margin-top: 20px;
            text-align: center;
            font-size: 1.5em;
            font-weight: bold;
            border: 2px solid #27ae60;
        }
        .stButton>button {
            background-color: #27ae60;
            color: white;
            border-radius: 8px;
            padding: 12px 24px;
            width: 100%;
            transition: all 0.3s;
            font-size: 1.1em;
        }
        .stButton>button:hover {
            background-color: #2ecc71;
            transform: scale(1.02);
        }
        .stSelectbox, .stNumberInput {
            margin-bottom: 15px;
        }
        .footer {
            text-align: center;
            margin-top: 30px;
            color: #777;
            font-size: 0.9em;
        }
    </style>
""", unsafe_allow_html=True)

# Load model with caching and error handling
@st.cache_resource
def load_model():
    try:
        with open("final_model_.pkl", "rb") as f:
            model = pickle.load(f)
        st.success("✅ Model loaded successfully!")
        return model
    except FileNotFoundError:
        st.error("❌ Model file not found! Please ensure 'final_model_.pkl' is in the correct directory.")
        return None
    except Exception as e:
        st.error(f"❌ Error loading model: {str(e)}")
        return None

model = load_model()

# App title and header
st.markdown("<h1 class='title'>⚡ Smart Electric Bill Predictor</h1>", unsafe_allow_html=True)
st.markdown("Predict your monthly electricity bill based on appliance usage patterns and location.")

# Main input section
with st.expander("🏠 **Property & Usage Details**", expanded=True):
    st.markdown("<div class='input-section'>", unsafe_allow_html=True)
    
    col1, col2 = st.columns(2)
    
    with col1:
        st.subheader("Appliance Usage (Hours)")
        fan = st.slider("Fan Hours per Day", 5.0, 23.0, 13.0, 0.5,
                       help="Daily usage hours of ceiling/table fans")
        fridge = st.slider("Refrigerator Hours", 17.0, 23.0, 18.0, 0.5,
                          help="Refrigerator running hours (typically 18-24 hours)")
        ac = st.slider("Air Conditioner Hours", 0.0, 3.0, 1.0, 0.5,
                      help="Daily AC usage hours (0 if not used)")
        tv = st.slider("Television Hours", 3.0, 22.0, 12.0, 0.5,
                      help="Daily TV viewing hours")
        monitor = st.slider("Computer Monitor Hours", 1.0, 12.0, 8.0, 0.5,
                           help="Daily computer usage hours")
        
    with col2:
        st.subheader("Location & Billing")
        city = st.selectbox("City", 
                          ['Hyderabad', 'Vadodara', 'Shimla', 'Mumbai', 'Ratnagiri',
                           'New Delhi', 'Dahej', 'Ahmedabad', 'Noida', 'Nagpur', 'Chennai',
                           'Faridabad', 'Kolkata', 'Pune', 'Gurgaon', 'Navi Mumbai'],
                          help="Select your city for regional tariff rates")
        
        company = st.selectbox("Electricity Provider", 
                              ['Tata Power Company Ltd.', 'NHPC', 'Jyoti Structure',
                               'Power Grid Corp', 'Ratnagiri Gas and Power Pvt. Ltd. (RGPPL)',
                               'Adani Power Ltd.', 'Kalpataru Power', 'Orient Green',
                               'Sterlite Power Transmission Ltd',
                               'Neueon Towers / Sujana Towers Ltd.', 'KEC International',
                               'Indowind Energy', 'Unitech Power Transmission Ltd.',
                               'Bonfiglioli Transmission Pvt. Ltd.', 'SJVN Ltd.',
                               'Maha Transco – Maharashtra State Electricity Transmission Co, Ltd.',
                               'L&T Transmission & Distribution', 'Guj Ind Power',
                               'Torrent Power Ltd.', 'Reliance Energy', 'GE T&D India Limited',
                               'NTPC Pvt. Ltd.',
                               'Optibelt Power Transmission India Private Limited', 'CESC',
                               'Ringfeder Power Transmission India Pvt. Ltd.', 'Reliance Power',
                               'JSW Energy Ltd.', 'Sunil Hitech Eng',
                               'Toshiba Transmission & Distribution Systems (India) Pvt. Ltd.',
                               'Jaiprakash Power', 'TransRail Lighting', 'NLC India'],
                              help="Select your electricity provider company")
        
        month = st.selectbox("Month", 
                            ["January", "February", "March", "April", "May", "June",
                             "July", "August", "September", "October", "November", "December"],
                            index=7,  # Default to August
                            help="Select month for seasonal variation")
        
        monthly_hours = st.slider("Total Monthly Usage Hours", 95.0, 926.0, 826.0, 10.0,
                                help="Sum of all appliance usage hours for the month")
        
        tariff_rate = st.slider("Electricity Tariff Rate (₹/kWh)", 7.4, 9.3, 8.2, 0.1,
                              help="Current electricity rate per unit")
    
    st.markdown("</div>", unsafe_allow_html=True)

# Prediction button and results
if st.button("🔍 Predict Monthly Bill", use_container_width=True):
    if model is None:
        st.error("Cannot make prediction - model not loaded.")
    else:
        try:
            # Convert month name to number
            month_num = ["January", "February", "March", "April", "May", "June",
                        "July", "August", "September", "October", "November", "December"].index(month) + 1
            
            input_data = pd.DataFrame([[fan, fridge, ac, tv, monitor, month_num, 
                                      city, company, monthly_hours, tariff_rate]],
                                    columns=['Fan', 'Refrigerator', 'AirConditioner', 
                                             'Television', 'Monitor', 'Month', 
                                             'City', 'Company', 'MonthlyHours', 
                                             'TariffRate'])
            
            predicted_price = model.predict(input_data)[0]
            
            # Display result with visual impact
            st.markdown(f"""
                <div class='result-box'>
                    <div style='font-size: 1.2em; margin-bottom: 10px;'>Estimated Monthly Bill</div>
                    <div style='font-size: 2em; color: #27ae60;'>₹ {predicted_price:,.2f} INR</div>
                    <div style='margin-top: 15px; font-size: 0.8em; color: #555;'>
                        Based on your usage patterns in {city} ({month})
                    </div>
                </div>
            """, unsafe_allow_html=True)
            
            # Add energy saving tips based on prediction
            if predicted_price > 5000:
                st.warning("💡 High Bill Alert: Consider reducing AC usage and switching to energy-efficient appliances.")
            elif predicted_price > 3000:
                st.info("💡 Moderate Bill: You might save by using fans instead of AC when possible.")
            else:
                st.success("💡 Efficient Usage: Your electricity consumption is well managed!")
            
        except Exception as e:
            st.error(f"❌ Prediction error: {str(e)}")

# Additional information section
with st.expander("ℹ️ About This Prediction"):
    st.markdown("""
    **How this prediction works:**
    - The model analyzes your appliance usage patterns, location, and local electricity rates
    - Calculations consider seasonal variations in energy consumption
    - Predictions are based on machine learning models trained on historical billing data
    
    **For more accurate results:**
    - Provide exact usage hours from your electricity meter if available
    - Update tariff rates according to your latest electricity bill
    - Consider seasonal adjustments for AC/heating usage
    """)

# Footer
st.markdown("""
    <div class='footer'>
        Energy Conservation Starts With Awareness • Powered by Machine Learning
    </div>
""", unsafe_allow_html=True)