Interactive Signal Processing
Unit and Integration tests with Dash
Cool Youtube Playlist about Automated CI/CD with GitHub actions - Videos are very long but content is really complete with many practical examples.
PyLint - Lint is a tool that checks code Styling such as Identation, Line length, File Length and so on.
If we decide to use Mittweida servers instead of our own computers, we can use those tutorials to set them up. I never used it so I am not sure about capabilities and Limitations
Website SetUp
Data Base SetUp
If we are going to make this app scalable, we need to decide on one architecture to use: Here is a small Medium article on Architecture with dash.
Clean Architecture for AI ML applications using dash and plotly with docker
I would recomend MVC for the UI since it is quite simple.
Honestly, I don't know if it is a thing in python, but basically what I mean is that we can access those variables without having to make an instance of this class. Global Variables or Singletons could also be used. I am trying to order these points by importance
File | Reasoning |
---|---|
routes | For type safety and also better to have an overview of all routes used in the app. Could be in root folder and each subfolder be a different route. TBD |
Strings | Think About Localization |
Styling | Think about using university colors and logos |
Maybe it would make sense to put all code inside a Docker Container, since we are planning to move the code to University Servers at some point. Docker Compose makes the migration almost effortless.
- We need to define some general rules for our branches. It is usually bad practice to have huge branches because it is a pain in the ass to solve merge conflicts.
- Branch naming
- Setup Branch Protection.