Unitesting for Data and Models

There are ways of kind of monitoring whether your model has drifted too much from, like, the last version to the new version. You can analyze the distribution of classes of your model's predictions. Unitesting for data and models is a great idea - it's almost effortless to do but could save you so much grief. And then i was going to ask about miteration cycles. So i'll just start again. Aodsit nois. Well, do the last part again. E, am, so yab, i prefer not to think about the for the future. I try to avoid the a robo take over stuff,. But i guess we'd be in a very

Transcript

Play full episode

Remember Everything You Learn from Podcasts

Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.