Skip to content

README.md

Latest commit

 

History

History
12 lines (9 loc) · 946 Bytes

README.md

File metadata and controls

12 lines (9 loc) · 946 Bytes
Raw

Life Expectancy Prediction Model using Flask and Heroku

Using ML model to predict Life Expectancy for each country and how to deploy a Python Flask application to Heroku.

3 options for running the project:

  1. Local run for testing - run as a python file server.py without Gunicorn locally. Since environment variable 'PORT' does not exist, will listen on default Flask port 5000 on localhost only.
  2. Heroku without Gunicorn - run as a python file server.py without Gunicorn on Heroku. Use the following line in Procfile: web: python server.py. Will listen on port given by Heroku in environment variable PORT
  3. Heroku with Gunicorn - change Procfile file to: web: gunicorn server:app. Then Gunicorn will run inside Heroku dyno with multiple instances of the server.py specified by Heroku in environment variable WEB_CONCURRENCY depending on available and used memory in the dyno.