abersbail/car-price-model

🚗 AI Car Price Prediction Model (Gradient Boosting)

This is a trained Gradient Boosting Regressor model designed for estimating used car market values based on key vehicle attributes.

📊 Model Performance

• R² Score: 0.9787 (97.87% variance explained)
• Mean Absolute Error (MAE): $2,390.93

🧮 Model Inputs

1. brand: Vehicle make (Toyota, Honda, Ford, BMW, Mercedes-Benz, Audi, Hyundai, Tesla, Chevrolet, Volkswagen)
2. year: Model year (2010 - 2025)
3. mileage: Total mileage driven
4. body_type: Sedan, SUV, Hatchback, Coupe, Truck, Convertible
5. fuel_type: Petrol, Diesel, Hybrid, Electric
6. transmission: Automatic, Manual
7. condition: Excellent, Good, Fair, Poor
8. engine_size: Engine displacement in Liters (1.0 - 5.0L)
9. horsepower: Engine power output in HP
10. owners: Number of previous owners (1 - 4)

💻 Python Usage Example

import pickle
import pandas as pd
from huggingface_hub import hf_hub_download

# Download model from Hugging Face Hub
model_path = hf_hub_download(repo_id="abersbail/car-price-model", filename="car_price_model.pkl")

with open(model_path, "rb") as f:
    model = pickle.load(f)

# Sample prediction
sample_car = pd.DataFrame([{
    'brand': 'Toyota',
    'year': 2022,
    'mileage': 25000,
    'body_type': 'SUV',
    'fuel_type': 'Hybrid',
    'transmission': 'Automatic',
    'condition': 'Excellent',
    'engine_size': 2.5,
    'horsepower': 219,
    'owners': 1
}])

estimated_price = model.predict(sample_car)[0]
print(f"Estimated Market Value: ${estimated_price:,.2f}")

📜 License

MIT License