*Reproducible Deep Learning*

The first two exercises are out!

We start quick and easily, with some simple manipulation on Git branches, scripting, audio classification, and configuration with @Hydra_Framework.

Small
thread with all information 🙃 /n

Reproducibility is associated to production environments and MLOps, but it is a major concern today also in the research community.

My biased introduction to the issue is here: https://t.co/PqWH6uL5eT
The local setup is on the repository: https://t.co/9mhtZoJhE9

The use case for the course is a small audio classification model trained on event detection with the awesome @PyTorchLightnin library.

Feel free to check the notebook if you are unfamiliar with the task. /n
I spent some time understanding how to make the course as modular and "reproducible" as possible.

My solution is to split each exercise into a separate Git branch containing all the instructions, and a separate branch with the solution.

Two branches for now (Git and Hydra). /n
How well do you *really* know Git? The more I learn, the more I find it incredible.

I summarized most of the information on a separate set of slides: https://t.co/6dSmK3IfWB

Be sure to check them out before continuing! /n
Exercise 1 is a simple example of turning a notebook into a working script.

To make things more interesting, you have to complete the exercise while working on a separate Git branch!

https://t.co/35y9NQaVtz

Nothing incredible, but it is always good to start small /n
Once you have a working training script, it is time to add some "bell and whistles"!

My must-have is some external configuration w/ @Hydra_Framework. Exercise 2 guides you in all the required steps.

Plus side: colored logging!

https://t.co/WyAIMXASRl
That is all for the moment. The next exercises will explore having external data versioning with @DVCorg and complete isolation with @Docker.

You can follow by starring the official repository, or here on Twitter. 👀

https://t.co/9mhtZoJhE9

More from All

You May Also Like

The YouTube algorithm that I helped build in 2011 still recommends the flat earth theory by the *hundreds of millions*. This investigation by @RawStory shows some of the real-life consequences of this badly designed AI.


This spring at SxSW, @SusanWojcicki promised "Wikipedia snippets" on debated videos. But they didn't put them on flat earth videos, and instead @YouTube is promoting merchandising such as "NASA lies - Never Trust a Snake". 2/


A few example of flat earth videos that were promoted by YouTube #today:
https://t.co/TumQiX2tlj 3/

https://t.co/uAORIJ5BYX 4/

https://t.co/yOGZ0pLfHG 5/