The current features aren't designed for deploying models only for training them. That's my understanding at least but it doesn't mean it can't be used for deploying. I've seen some sources point to Core ML for productionizing models, but I haven't looked into it enough to know how that works.
Why is it a good toolkit in research and not production?
The current features aren't designed for deploying models only for training them. That's my understanding at least but it doesn't mean it can't be used for deploying. I've seen some sources point to Core ML for productionizing models, but I haven't looked into it enough to know how that works.
In any case im not sure why would anyone want to deploy on apple products , except maybe integrated local AIs, which aren’t here yet
Making anything usable on Apple devices will require it. That's a huge consumer pool.