🚗 Car Price Prediction with Machine Learning

URL: Webpage

Welcome to the Car Price Prediction project! This repository provides a complete pipeline for predicting used car prices using machine learning, including data cleaning, visualization, model training, API deployment, and a user-friendly Streamlit web app.

---

📂 Project Structure

CAR-PRICE-PREDICTION-WITH-MACHINE-LEARNING/
│
├── Api/                # FastAPI backend for model inference
│   └── main.py
│
├──Models/
│   ├── linear_regression_model.pkl
│   ├── random_forest_model.pkl
│   └── xgboost_model.pkl
│
├── Streamlit_app/      # Streamlit frontend for user interaction
│   └── app.py
│
├── Data/               # Datasets
│   ├── car_data.csv
│   └── cleaned_car_data.csv
│
├── Notebooks/          # Jupyter notebooks for EDA, cleaning, training
│   ├── data_clean.ipynb
│   ├── data_visualization.ipynb
│   └── model_training.ipynb
│   
├── requirements.txt
├── render.yaml         # (Optional) Render deployment config
└── README.md

---

🚀 Quick Start

1. Clone the Repository

git clone https://github.com/your-username/CAR-PRICE-PREDICTION-WITH-MACHINE-LEARNING.git
cd CAR-PRICE-PREDICTION-WITH-MACHINE-LEARNING

2. Install Dependencies

Install Python 3.12+ and pip, then:

pip install -r requirements.txt

3. Run the FastAPI Backend

cd Api
uvicorn main:app --reload --host 0.0.0.0 --port 8000

4. Run the Streamlit Frontend

Open a new terminal:

cd Streamlit_app
streamlit run app.py

5. Use the App

• Go to http://localhost:8501 in your browser.
• Enter car details and get instant price predictions from multiple models.
• See feature importances and compare model performances.

---

🧑‍💻 Features

• Data Cleaning: Remove outliers, filter non-car entries, handle missing values.
• Visualization: Explore data distributions and relationships with Seaborn and Matplotlib.
• Model Training: Train and compare Linear Regression, Random Forest, and XGBoost models.
• API: FastAPI backend serves predictions and model confidences.
• Frontend: Streamlit app with car name suggestions, input validation, and attractive UI.
• Model Selection: Automatically selects the best model based on confidence (R² score).
• Deployment Ready: Easily deploy both backend and frontend on Render or similar platforms.

---

📊 Data Columns

• Car_Name: Name of the car (with suggestions in the UI)
• Year: Year of purchase
• Selling_Price: Price at which the car is being sold (target)
• Present_Price: Current ex-showroom price
• Driven_kms: Kilometers driven
• Fuel_Type: Petrol, Diesel, CNG, etc.
• Selling_type: Dealer or Individual
• Transmission: Manual or Automatic
• Owner: Number of previous owners

---

🏆 Model Performance

Model	MAE	MSE	R² Score (%)
Linear Regression	1.36	3.1	92.44
Random Forest	1.19	3.1	92.43
XGBoost	1.58	5.6	86.35

Random Forest is selected as the best model by default.

---

🌐 Deployment on Render

1. Push your code to GitHub.
2. Create two Render web services:
   • One for the FastAPI backend (Api/)
   • One for the Streamlit frontend (Streamlit_app/)
3. Set the correct start commands:
   • FastAPI: uvicorn main:app --host 0.0.0.0 --port 10000
   • Streamlit: streamlit run app.py --server.port 10001 --server.address 0.0.0.0
4. Set environment variables if needed (e.g., API URL for frontend).
5. See render.yaml for an example configuration.

---

📒 Notebooks

• data_clean.ipynb: Data cleaning, outlier detection (IQR method), and preprocessing.
• data_visualization.ipynb: Distribution plots, scatter plots, and correlation heatmaps.
• model_training.ipynb: Model training, evaluation, and feature importance plots.

---

🤝 Contributing

Contributions are welcome! Please open an issue or submit a pull request.

---

📢 Acknowledgements

• Streamlit for the frontend.
• FastAPI for the backend.
• scikit-learn, XGBoost for machine learning.
• Render for easy deployment.

---

📬 Contact

For questions or suggestions, please open an issue or contact the maintainer.

---

Happy Predicting! 🚗