Grap vs. Grafts in Deep Learning?

Grafts seemed like the obvious choice. Not having a graph meant, ok, how do you deply now? So that was probably what selted the balance for us and eventually we ended up with a graph. We are in just the right place to push on thaten and leverage thaten and deliver on lots of things that people want. And i would say this kind of growth probably started seeing somewhat after the open sourcing,. where like, ok, your deep learning is actually growing awar faster for a lot of different reasons.