bakaoh / GiraffeTools

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Website: www.giraffe.tools

Giraffe Tools

This web application is in the early stages of construction

Giraffe

a Graphical Interface for Reproducible Analysis oF workFlow Experiments

This is a web application with a set of tools to build and improve your data analysis! Initially, this will focus on neuroscientific applications. The first goal is to make a web application from my earlier project Porcupine, a visual workflow editor. This can further be developped to support version control of a workflow by means of Github integration, connect to visualisation or execution platforms, and much more.

Intended usage

The plan is to have a user go to: https://www.giraffe.tools/$username/$repository/$branch and there find a dashboard of the project. A project is a GitHub repository that is characterised by a GIRAFFE.yml configuration file in its root and links to configuration files of specific tools. The rest of the repository doesn't matter.

This doesn't work yet: Example: https://www.giraffe.tools/TimVanMourik/SomeGiraffeExample/master

This is similar in usage to for example GitPitch, which is how the exaplantory presentation about this repository was made: https://gitpitch.com/TimVanMourik/GiraffeTools/master.

Potential Tools

  • Porcupine (pipeline creator), largely based on a similar implementation for deep learning
    • Must have: Create analysis code from workflow
    • Stretch goal: Create paper snippets from workflow
  • OAuth Github link to easily interface with projects
  • Analysis preregistration
  • DOI (static link) for analysis
  • Visualisation of the data that flows through pipeline
    • Stretch goal: Augmented Reality visualisation (like this project). This should definitely be called ARmadillo
  • Code execution integration, via, e.g., Amazon
  • [Your input here!]

Interesting links:

General

  • This website is currently deployed on Heroku with a Django + Node.js build. The Docker image has a (near) identical setup. You can run this web application locally by installing and running Docker, and simply typing docker-compose up in the terminal/command prompt.
  • Join us on Slack!

About

License:GNU General Public License v3.0


Languages

Language:JavaScript 47.4%Language:CSS 31.8%Language:Python 13.7%Language:HTML 6.4%Language:Shell 0.6%