app.py

import streamlit as st
import joblib
import pandas as pd

# Load the trained model
model_path = "random_forest_model.pkl"
model = joblib.load(model_path)

# Title of the web app
st.title("Car Price Prediction")

# Input fields for user
brand = st.text_input("Brand", "BMW")
model_name = st.text_input("Model", "320i")
year = st.number_input("Year", min_value=2000, max_value=2024, value=2018)
mileage = st.number_input("Mileage (in km)", min_value=0, max_value=500000, value=50000)
fuel_type = st.selectbox("Fuel Type", ["Petrol", "Diesel", "Hybrid", "Electric"])
engine_size = st.number_input("Engine Size (in liters)", min_value=0.5, max_value=6.0, value=2.0)

# Prediction function
def predict_price(brand, model_name, year, mileage, fuel_type, engine_size):
    # Create a DataFrame for input
    input_data = pd.DataFrame([[brand, model_name, year, mileage, fuel_type, engine_size]],
                              columns=["Brand", "Model", "Year", "Mileage", "Fuel Type", "Engine Size"])
    
    # Make prediction
    prediction = model.predict(input_data)
    
    return f"Estimated Price: {prediction[0]:,.2f} €"

# Button to make a prediction
if st.button("Predict Price"):
    result = predict_price(brand, model_name, year, mileage, fuel_type, engine_size)
    st.success(result)