Ein Workshop von Sandro Braun, Leander Kurscheidt auf der GPN19.
Modern ML workflow requires to run experiments fast at a large scale. In order to stay sane and keep an overview of what is going on, there are some tools out there.
We will compare 3 tools in this workshop : ML flow, sacred and edflow. These tools focus on different parts of the development workflow : fast model iteration, monitoring (dashboard) and deployment.
We will provide some basic code and instructions on how to add the tools to basic examples. We will cover standard classification based problems (i.e. getting the largest number on some problem) and image based problems (i.e. GANs, generative models, etc...)
Anything will be available on Github afterwards.
https://mlflow.org/ https://github.com/IDSIA/sacred https://github.com/pesser/edflow