How to effectively test and debug machine learning models, from ML engineer@Apple to startup founder - Gabriel Bayomi - the data scientist show #055
The Data Scientist Show - Daliana Liu
00:00
How to Fix an Error in a Model?
The idea is like after, you know, identify a problem and try to fix it. And then the best thing I might create is that on the subpopulation, you want to make sure precision is higher than 0.9 or something like that. So we also show you, Hey, if this new version of your model, there's like these examples that were regression. And here's why showing also the accessibility for each one of them. It doesn't tell you like it's not actionable, right? There's no like what should I do next? A lot of what we built was like on this action part of like find out this list of data doesn't work, take an
Transcript
Play full episode
