TenserFlow

The downside of so many people being excited about tenser flow and becoming to rely on it in many other applications is that you're kind of responsible. You're responsible for previous versions to some degree, still working. And yes, it does come a huge cost. There's a we have to think about a lot of things as we do new things and make new changes. Thata two point o does break some bak compatibility, but not too much. It seems like the conversionis pretty straightforward. A do do you think that's still important, given how quickly deep learning is changing? Can you just start over? Or is there pressure to naught?