The backend and Front end is managed by Dash framework, encapsulated in a general server python application (Flask).

Github

The general architecture is composed by some services: Front-end: We use Dash Framework to generate the interactive web-ui with Python . Back-end: The Dash App is encapsulated in a Flask Server (The main idea is to use a general server, which lets us scale the project). Database: RDS (POSTGRESQL) only accessible by EC2

scikit-learn flask dash pandas numpy sqlalchemy

FLASK_APP=data4all.py ENV_PATH=app/ DB_PASSWORD=The db password

• app

  • __init__.py: code for serving application
  • routes: contains example for general api using flask
  • assets: images
  • data: csv and json files
  • libpage
    • content.py: general components
    • map_cluster.py: map component layout
    • modal_info.py: modal component to show information
    • ui_inputs: inputs components
  • dashboard -utils: utilities like Database Connection, reading files, etc -locator_dashboard: handle information and callbacks between ui and cluster algorithm

• data4all.py: the main file (you can use flask run or gunicorn)

The Web application is deployed in AWS. In addition for the general configuration, production environment needs some configuration: Web Serving and Reverse Proxying: We use Nginx for web serving and load balancing. Wsgi Server: We use Gunicorn in order to handle simultaneous requests. The number of workers is configured to four.