• Note: hotfix for werkzeug issue #1992 in place
• Note: changed deployment from Heroku to Google Cloud Run

Docker Dash Example

A simple design for a plotly-dash app with sklearn running within a docker container deployed to Google Cloud Run using CI/CD.

For a deep dive on the implementation please see:

1. Initial deployment through Heroku.
2. Current deployment to GCP through Google Cloud Run

Inspired by This TDD Course

Using pre-commit-hooks

• flake8
• black
• isort
• detect-private-key
• bandit

Repo structure

├── .github
│   └── workflows
│        └── main.yml
│
├── project
│   ├── app
│   │   ├── __init__.py
│   │   ├── app.py
│   │   ├── functions.py
│   │   └── assets
│   ├── tests
│   │   └── test_functions.py
│   ├── Dockerfile
│   ├── Dockerfile.prod
│   └── requirements.txt
│
├── release.sh
├── setup.cfg
├── .pre-commit-config.yaml
├── .gitignore
│
└── README.md

Run locally

To run the image locally, cd into the docker-dash-example folder and:

docker build -t docker-dash project/.

And run the container

docker run -p 8050:8050 docker-dash

You can find to the app on your local machine http://localhost:8050/ (or localhost:8050). This way the image is created using the Dockerfile, instead of the Dockerfile.prod.

Project context

This was a fun project to apply different learnings in practice, very happy to hear your thoughts in the repo, including feedback of course!