Why you shouldn't use K-Fold Cross Validation.
10 Fold cross validation isn't a silver bullet. Your model can still generalize poorly or heavily overfit the data even with a CV scheme! Here I'll show an example as to why!
Table of Content
Introduction: 0:00
Paper Walkthrough: 1:45
DEAP Dataset: 2:23
Issue: 3:09
Temporal Correlation: 5:35
Rant: 9:11
Conclusion: 10:17
My point here is not to diss the paper I've outlined, but more to pinpoint a trend I've seen a bit too often where X is claim while the machine learning model haven't really learned it.
It's like the 3rd article that claim huge numbers for emotion recognition with the DEAP dataset too so its a bit annoying, especially having worked on that problem for a while.
----
Join the Discord for general discussion: / discord
----
Follow Me Online Here:
Twitter: / codethiscodeth1
GitHub: https://github.com/yacineMahdid
LinkedIn: / yacine-mahdid-809425163
Instagram: / yacine_mahdid
___
Have a great week! 👋