/house-price-prediction

Primary LanguagePython

🏡 House Price Prediction

This repository contains a Python script that generates a model, stores it in a file and then uses it to generate a house price prediction based on the property parameters.

Stack

  • Backend = Flask
  • Frontend = React with use of Material UI
  • DB = sqlite (on startup is created at flask-backend/instance/data.db)

How to run locally

Use docker-compose up to run backend and frontend together in one container with all requirements.

docker-compose up
# And open http://127.0.0.1:5000

Possible future improvements

Backend:

  • Cache the requests in Redis
  • Use Gunicorn or some WSGI
  • Take care of CORS
  • Add tests for the sample inputs
  • Add versioning to the API + docs endpoint
  • What should happen with duplicate requests? So far it is just saved with new timestamp

Frontend:

  • Refine the details - favicon, title etc.
  • Cypress tests
  • Check how it looks on smaller devices
  • Add clear tabe option

General:

  • Add linter + pre-commit