Predictor

A web app that predicts used car prices using a linear regression model. Input features include company, model, year, fuel type, and kilometers driven. Built with Flask and scikit-learn, with a REST API and Docker support.

Docker Setup

docker build -t car-price-predictor .
docker run -p 5000:5000 car-price-predictor

Access the app at http://localhost:5000. Ensure LinearRegressionModel.pkl and Cleaned_Car_data.csv are in the project root.

Overview

Data was cleaned and analysed and outliers were also removed. Model had R2 score of 0.89. You can see CarPricePrediction.ipynb for cleaning of data and training the model details.

## API Endpoint
POST /predict

Form data parameters:
- company (string)
- model (string)
- year (integer)
- fuel_type (string)
- kms (integer)

Response:
{
    "prediction": 450000.0
}

---