Is Your Model Good Enough?

Getting a sense of whether or not the model's actually good enough to wa what you need to use it for. Um, usually that requires a kind of input from some stake holder. You can't really automate like, oh, you know, my model has 99 % accuracy, therefore it's good enough. Like, that's not guaranteed. It depends on the task. Ah, and then the the decision to maybe deploy, or to switch the main model with a new model, or something like that. That still seems to me as well, like, speaking to companies. A, it's something people would want to automate, but i haven't heard of anyone successfully doing that.

Transcript

Play full episode

Remember Everything You Learn from Podcasts

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