🚗 Car Price Predictor

An intelligent car price prediction application using Gradient Boosted Trees machine learning algorithm. This project features a beautiful Streamlit web interface for predicting car prices based on comprehensive vehicle specifications.

---

📋 Table of Contents

• Overview
• Features
• Demo
• Tech Stack
• Project Structure
• Installation
• Usage
• Model Details
• API Documentation
• Screenshots
• Future Enhancements
• Contributing
• License
• Contact

---

🎯 Overview

This machine learning project predicts the market value of cars based on 74 comprehensive features including vehicle specifications, engine details, and body characteristics. The model uses Gradient Boosting regression to provide accurate price estimates with confidence intervals.

Key Highlights

• ✅ 91% Prediction Accuracy (R² score)
• ✅ 74 Features analyzed (15 numerical + 59 categorical)
• ✅ Real-time Predictions with interactive UI
• ✅ One-Hot Encoding for categorical variables
• ✅ No Feature Scaling required (tree-based model)

---

✨ Features

🤖 Machine Learning

• Gradient Boosted Tree regression model
• Handles both numerical and categorical data
• One-hot encoding for categorical features
• High accuracy predictions (~91% R²)
• Robust to outliers and missing patterns

🎨 User Interface

• Beautiful Streamlit Dashboard with custom CSS
• Smart Price Formatting ($10.9K instead of $10,911.14)
• Price Range Estimation (±10% confidence interval)
• Quick Presets - Load example cars instantly:
  • Toyota Camry (Mid-range sedan)
  • BMW 3-Series (Luxury sport)
  • Honda Civic (Economy hatchback)
  • Mercedes E-Class (Premium sedan)
  • VW GTI (Sport turbo)
• Gradient Designs with modern UI/UX
• Real-time Validation for input fields
• Key Metrics Dashboard showing HP, MPG, Engine Size
• Responsive Layout for all screen sizes

📊 Predictions Include

• Estimated market value
• Exact price prediction
• Typical price range (lower/upper bounds)
• Key vehicle metrics summary
• Model confidence indicators

---

🎬 Demo

Try it yourself:

# Clone the repository
git clone https://github.com/Deathblu/car-price-predictor.git
cd car-price-predictor

# Install dependencies
pip install -r requirements.txt

# Run the app
streamlit run streamlit_app.py

Live Demo: [Deploy on Streamlit Cloud] (Coming Soon)

---

🛠️ Tech Stack

Machine Learning & Data Science

• Python 3.8+ - Core programming language
• scikit-learn - Gradient Boosting Regressor
• NumPy - Numerical computations
• Pandas - Data manipulation (during training)
• Joblib - Model serialization

Web Application

• Streamlit - Interactive web interface
• Custom CSS - Enhanced UI/UX with gradients

Development Tools

• Jupyter Notebook - Model training and experimentation
• Git - Version control

---

📁 Project Structure

car-price-predictor/
│
├── streamlit_app.py          # Main Streamlit application
├── final_model.pkl           # Trained Gradient Boosting model
├── requirements.txt          # Python dependencies
├── README.md                 # Project documentation
│
├── notebooks/                # (Optional) Jupyter notebooks
│   └── model_selection.ipynb  # Model development & training
│
├── data/                     # (Optional) Dataset folder
│   └── car_price.csv          # Original dataset
│
└── .gitignore               # Git ignore file

---

🚀 Installation

Prerequisites

• Python 3.8 or higher
• pip package manager

Step-by-Step Setup

1. Clone the repository

git clone https://github.com/Deathblu/car-price-predictor.git
cd car-price-predictor

2. Create virtual environment (recommended)

# Windows
python -m venv venv
venv\Scripts\activate

# macOS/Linux
python3 -m venv venv
source venv/bin/activate

3. Install dependencies

pip install -r requirements.txt

4. Verify installation

python -c "import streamlit; import joblib; import numpy; print('All dependencies installed!')"

---

💻 Usage

Running the Application

streamlit run streamlit_app.py

The app will automatically open in your browser at http://localhost:8501

Making Predictions

1. Choose a preset (optional) from the sidebar:

   • Select from Toyota Camry, BMW 3-Series, Honda Civic, etc.
   • Or enter custom values

2. Enter vehicle specifications:

   • Numerical features: symboling, dimensions, weight, engine specs, etc.
   • Categorical features: fuel type, transmission, body style, etc.

3. Click "Predict Price Now" button

4. View results:

   • Estimated market value
   • Exact prediction
   • Price range
   • Key metrics summary

Example Prediction

Input: 2018 Toyota Camry

• Engine: 2.0L (1998cc), 140 HP
• Fuel: Gas, MPFI
• Body: Sedan, 4 doors
• Transmission: Standard, FWD
• Fuel Economy: 24 city / 30 highway MPG

Output: ~$18,500 (range: $16,650 - $20,350)

---

🧠 Model Details

Algorithm

Gradient Boosted Trees (Gradient Boosting Regressor)

Why Gradient Boosting?

• ✅ Handles non-linear relationships
• ✅ Works with mixed data types
• ✅ No feature scaling required
• ✅ Robust to outliers
• ✅ High accuracy with proper tuning

Features Breakdown

Numerical Features (15)

1. symboling - Insurance risk rating
2. normalized_losses - Normalized loss value
3. drive_wheels - Drive wheels configuration
4. length - Vehicle length (inches)
5. width - Vehicle width (inches)
6. height - Vehicle height (inches)
7. curb_weight - Weight (lbs)
8. engine_size - Engine displacement (cc)
9. bore - Engine bore
10. stroke - Engine stroke
11. compression_ratio - Compression ratio
12. horsepower - Engine power (bhp)
13. peak_rpm - Peak RPM
14. city_mpg - City fuel economy
15. highway_mpg - Highway fuel economy

Categorical Features (One-Hot Encoded = 59 features)

• Fuel Type (2): diesel, gas
• Make (2): std, turbo
• Doors (2): two, four
• Aspiration (5): convertible, hatchback, sedan, wagon, hardtop
• Wheel Base (3): rwd, fwd, 4wd
• Engine Location (2): front, rear
• Body Style (22): alfa-romero, audi, bmw, chevrolet, dodge, honda, isuzu, jaguar, mazda, mercedes-benz, mercury, mitsubishi, nissan, peugot, plymouth, porsche, renault, saab, subaru, toyota, volkswagen, volvo
• Engine Type (6): dohc, ohcv, ohc, l, rotor, ohcf
• Cylinders (7): four, six, five, three, twelve, two, eight
• Fuel System (8): mpfi, 2bbl, mfi, 1bbl, spfi, 4bbl, idi, spdi

Total Features: 15 + 59 = 74 features

Model Performance

• R² Score: ~0.91 (91% accuracy)
• Training Method: Jupyter Notebook
• Serialization: Joblib (.pkl format)

Training Process

# Example training code structure
from sklearn.ensemble import GradientBoostingRegressor
import pandas as pd
import joblib

# Load and preprocess data
df = pd.read_csv('car_pricedata.csv')
X = pd.get_dummies(df.drop('price', axis=1))  # One-hot encoding
y = df['price']

# Train model
model = GradientBoostingRegressor(
    n_estimators=100,
    learning_rate=0.1,
    max_depth=5,
    random_state=42
)
model.fit(X, y)

# Save model
joblib.dump(model, 'final_model.pkl')

---

📡 API Documentation

Flask Backend (Optional)

If you need to integrate the model as an API:

Endpoint: POST /predict

Request:

{
  "features": [
    1, 120, 88.6, 168.8, 64.1, 48.8, 2548, 130, 3.47, 2.68, 9.0, 111, 5000, 21, 27,
    0, 1, 1, 0, 1, 0, 0, 0, 0, 1, 0, 0, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 1, 0, 0, 0,
    0, 0, 0, 0, 0, 1, 0, 0, 0, 0
  ]
}

Response:

{
  "success": true,
  "predicted_price": 18456.78
}

---

📸 Screenshots

Main Interface

Interactive dashboard with numerical and categorical input fields. Clean, modern design with gradient accents and organized feature sections.

Prediction Results

Beautiful results page showing estimated price in large gradient text ($10.9K format), exact value, price range (±10%), and key metrics cards for horsepower, fuel economy, and engine size.

Quick Presets

Sidebar with preset car examples (Toyota Camry, BMW 3-Series, Honda Civic, etc.) for quick testing and demonstration of the model.

---

🔮 Future Enhancements

Planned Features

• Historical Price Trends - Show price changes over time
• Comparison Tool - Compare multiple cars side-by-side
• Image Upload - Predict from car images using CNN
• Market Analysis - Show similar cars and market trends
• Export Reports - Download PDF price reports
• API Access - RESTful API for integrations
• Mobile App - Native iOS/Android applications
• Real-time Data - Integration with live market data

---

🤝 Contributing

Contributions are welcome! Here's how you can help:

1. Fork the repository
2. Create feature branch (git checkout -b feature/AmazingFeature)
3. Commit changes (git commit -m 'Add AmazingFeature')
4. Push to branch (git push origin feature/AmazingFeature)
5. Open Pull Request

Guidelines

• Follow PEP 8 style guide
• Add comments for complex logic
• Update documentation for new features
• Test thoroughly before submitting

---

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

---

👨‍💻 Contact

Your Name

• GitHub: @Deathblu
• LinkedIn: [(https://www.linkedin.com/in/anurag-wanwe-17b8ab243/)]
• Email: [anuwanwe1463@gmail.com]

Project Link: https://github.com/yourusername/car-price-predictor

---

🙏 Acknowledgments

• Dataset source: [UCI Machine Learning Repository / Kaggle]
• Streamlit team for the amazing framework
• scikit-learn community for excellent ML tools
• Stack Overflow community for troubleshooting help

---

📊 Project Stats

---

⭐ Star this repo if you find it helpful!

Made with ❤️ and 🐍 Python