app.py

import streamlit as st
import joblib
import numpy as np
import pandas as pd
st.markdown(
    """
    <style>
    body {
        background-color: #008080;
    }
    .stApp {
        background-color: #00FFFF;
    }
    h1 {
        text-align: center;
        color: #808080; /* Dark Red */
    }
    label, p, div {
        color: black !important;
    }
    .result {
        font-size: 24px;
        font-weight: bold;
        color: #808000; /* Red */
        text-align: center;
    }
    /* Style for button */
    div.stButton > button {
        background-color: #b71c1c !important; /* Dark Red */
        color: white !important; /* White Text */
        border-radius: 8px;
        padding: 10px 20px;
        font-size: 16px;
    }
    </style>
    """,
    unsafe_allow_html=True,
)



# Load trained model
model = joblib.load("carr.pkl")  # Make sure this file exists

# Streamlit UI
st.markdown("<h1 style='text-align: center; color: #1a237e;'> 🚗 Car Price Prediction 🚗</h1>", unsafe_allow_html=True)

# User Inputs
ID = st.text_input("ID") 
Levy = st.number_input("Levy", min_value=0, step=1)
Manufacturer = st.text_input("Manufacturer")
Model = st.text_input("Model")
Prod_year = st.number_input("Production Year", min_value=1900, max_value=2025, step=1)
Category = st.selectbox("Category", ["Sedan", "SUV", "Hatchback", "Coupe", "Convertible", "Pickup", "Minivan", "Other"])
Leather_interior = st.radio("Leather Interior", ["Yes", "No"])
Fuel_type = st.selectbox("Fuel Type", ["Petrol", "Diesel", "Electric", "Hybrid", "Other"])
Engine_volume = st.number_input("Engine Volume (L)", min_value=0.0, step=0.1)
Mileage = st.number_input("Mileage (km)", min_value=0, step=1000)
Cylinders = st.number_input("Cylinders", min_value=1, max_value=16, step=1)
Gear_box_type = st.selectbox("Gearbox Type", ["Manual", "Automatic", "CVT", "Other"])
Drive_wheels = st.selectbox("Drive Wheels", ["Front", "Rear", "All Wheel Drive"])
Doors = st.number_input("Doors", min_value=2, max_value=5, step=1)
Wheel = st.selectbox("Wheel Position", ["Left", "Right"])
Color = st.color_picker("Color")  # Might need conversion
Airbags = st.number_input("Airbags", min_value=0, max_value=12, step=1)

# Convert categorical values to numerical format
Leather_interior = 1 if Leather_interior == "Yes" else 0
Wheel = 1 if Wheel == "Left" else 0  # Convert to binary

# Create DataFrame
input_df = pd.DataFrame([[
    ID, Levy, Manufacturer, Model, Prod_year, Category, Leather_interior,
    Fuel_type, Engine_volume, Mileage, Cylinders, Gear_box_type, Drive_wheels,
    Doors, Wheel, Color, Airbags
]], columns=[
    "ID","Levy", "Manufacturer", "Model", "Prod_year", "Category", "Leather_interior",
    "Fuel_type", "Engine_volume", "Mileage", "Cylinders", "Gear_box_type",
    "Drive_wheels", "Doors", "Wheel", "Color", "Airbags"
])

# Prediction Button
if st.button("Predict Price 💰"):
    try:
        # Check input shape
        expected_features = model.n_features_in_
        if input_df.shape[1] != expected_features:
            st.error(f"Model expects {expected_features} features, but received {input_df.shape[1]}. Check input data.")
        else:
            prediction = model.predict(input_df)
            formatted_price = f"₹ {prediction[0]:,.2f}"
            st.success(f"Predicted Price: {formatted_price}")

    except Exception as e:
        st.error(f"Prediction Error: {str(e)}")