app.py

import streamlit as st
import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler, OneHotEncoder
from sklearn.compose import ColumnTransformer
from sklearn.pipeline import Pipeline
from sklearn.ensemble import RandomForestRegressor
from sklearn.metrics import mean_squared_error

# Load and prepare the dataset
url = "https://raw.githubusercontent.com/manishkr1754/CarDekho_Used_Car_Price_Prediction/main/notebooks/data/cardekho_dataset.csv"
df = pd.read_csv(url)

# Data preparation
df = df.drop(columns=['Unnamed: 0'])  # Drop irrelevant column
X = df.drop(columns=['selling_price'])
y = df['selling_price']

# Define feature types
num_features = ['vehicle_age', 'km_driven', 'mileage', 'engine', 'max_power', 'seats']
cat_features = ['car_name', 'brand', 'model', 'seller_type', 'fuel_type', 'transmission_type']

# Preprocessing pipeline
numeric_transformer = StandardScaler()
onehot_transformer = OneHotEncoder(handle_unknown='ignore')

preprocessor = ColumnTransformer(
    transformers=[
        ('num', numeric_transformer, num_features),
        ('cat', onehot_transformer, cat_features)
    ])

# Combine preprocessing with model
model = Pipeline(steps=[
    ('preprocessor', preprocessor),
    ('regressor', RandomForestRegressor(n_estimators=100, random_state=42))
])

# Split the data and train the model
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
model.fit(X_train, y_train)

# Evaluate the model (optional, you can remove this if not needed)
y_pred = model.predict(X_test)
mse = mean_squared_error(y_test, y_pred)

# Streamlit app
st.title('Used Car Price Prediction')

# Input fields
st.sidebar.header('Enter Car Details')

year = st.sidebar.number_input('Year of Manufacture', min_value=1990, max_value=2023, value=2015)
km_driven = st.sidebar.number_input('Kilometers Driven', min_value=0, max_value=300000, value=50000)
vehicle_age = st.sidebar.number_input('Vehicle Age (years)', min_value=0, max_value=30, value=5)
mileage = st.sidebar.number_input('Mileage (km/l)', min_value=0.0, max_value=50.0, value=15.0)
engine = st.sidebar.number_input('Engine Capacity (cc)', min_value=0, max_value=5000, value=1500)
max_power = st.sidebar.number_input('Maximum Power (bhp)', min_value=0, max_value=500, value=100)
seats = st.sidebar.number_input('Number of Seats', min_value=2, max_value=7, value=5)

seller_type = st.sidebar.selectbox('Seller Type', ['Dealer', 'Individual'])
transmission_type = st.sidebar.selectbox('Transmission Type', ['Manual', 'Automatic'])
fuel_type = st.sidebar.selectbox('Fuel Type', ['Petrol', 'Diesel', 'CNG', 'LPG'])
car_name = st.sidebar.text_input('Car Name')
brand = st.sidebar.text_input('Brand')
model_name = st.sidebar.text_input('Model')

# Button to trigger the prediction
if st.sidebar.button('Predict Price'):
    # Create input dataframe
    input_data = pd.DataFrame({
        'vehicle_age': [vehicle_age],
        'km_driven': [km_driven],
        'mileage': [mileage],
        'engine': [engine],
        'max_power': [max_power],
        'seats': [seats],
        'car_name': [car_name],
        'brand': [brand],
        'model': [model_name],
        'seller_type': [seller_type],
        'fuel_type': [fuel_type],
        'transmission_type': [transmission_type]
    })

    # Predict the price
    predicted_price = model.predict(input_data)
    st.write(f'The predicted selling price for the car is: ₹ {predicted_price[0]:,.2f}')