Used Car Price Prediction

This project is part of a larger initiative I’m working on for my contracted role as a data scientist. The goal? To build a machine learning model that predicts the market price of used cars and helps identify the best car listing deals in Latvia. 🚘

The dataset used in this project spans from 2018 to 2023 and is part of the Latvian Used-car Market Announcements Monitoring initiative, ran by Valerijs Skribans, Aivars Gulbis, and Aivars Krastins. The monitoring began in 2018 in response to Germany's Federal Administrative Court decision allowing diesel car bans to reduce air pollution.

Using this dataset, I trained an XGBoost model and achieved an R² score of 0.96, demonstrating the model's high accuracy in predicting used car prices.

For this milestone, I deployed my first ML-powered web service using Flask, Docker, GitHub Actions for CI/CD and Google Cloud Run. From handling features like mileage and fuel type to finding the best car deals, this project gave me real-world, hands-on experience building, testing and deploying machine learning models in production! 🌟