There’s two ways:

Using a Procfile in your repository root:

web: gunicorn --worker-tmp-dir /dev/shm myproject:app

Or, specify a custom run command found under “Components > Commands” tab for your app.

Also see @kamaln7’s post here: https://www.digitalocean.com/community/questions/configure-a-build-command-for-a-python-project-without-using-the-web-ui

Check this one
https://www.digitalocean.com/community/tutorials/how-to-serve-flask-applications-with-uswgi-and-nginx-on-ubuntu-18-04

Per the official Flask recommendation, you should use a WSGI server to run your app. Once configured and added to requirements.txt or Pipfile, set your app’s run command according to the WSGI server’s usage. For example, with Waitress:

waitress-serve --call 'flaskr:create_app'

Or with Gunicorn:

gunicorn --worker-tmp-dir /dev/shm myproject:app