Car Price Prediction Project

This project is a web application designed to predict car prices based on various input features using machine learning models. The application is developed using FastAPI for the API framework and Scikit-learn for model training and evaluation.

Features

• Provides car price predictions based on user-provided features.
• Exposes a REST API using FastAPI for seamless integration.
• Utilizes multiple machine learning models to enhance prediction accuracy.

Requirements

• Python 3.7 or higher
• FastAPI
• Uvicorn
• Pandas
• NumPy
• Scikit-learn

Project Workflow

1. Data Collection and Preprocessing: Collect and preprocess data, including features such as car model, year, mileage, engine size, and other relevant attributes. Handle missing values and encode categorical data.
2. Model Training: Train various machine learning models using the Scikit-learn library. Evaluate models using metrics such as Mean Absolute Error (MAE), Mean Squared Error (MSE), and R-squared values to determine the best-performing model.
3. API Development: Develop RESTful API endpoints using FastAPI to serve the model predictions. Implement endpoints for predicting car prices and retrieving model details.
4. Testing: Perform unit and integration testing to ensure the API is functioning as expected and that the model predictions are accurate.
5. Deployment: Deploy the FastAPI application using Uvicorn. The API will be accessible via a web server for user interaction and integration with other applications.

Usage

• Start the FastAPI server to access the application.

• Use the /predict endpoint to send a POST request with car feature details to receive a price prediction.

• Example request:

  {
      "model": "Toyota Camry",
      "year": 2019,
      "mileage": 30000,
      "engine_size": 2.5,
      "fuel_type": "Petrol",
      "transmission": "Automatic"
  }

• The response will include the predicted price based on the provided car features.

Acknowledgements

• Special thanks to all contributors and the open-source community for the libraries and tools used in this project.