[swampthinker]

If I were to start a theoretical computer vision company today, which would I be better off using?

[probably_wrong]

My very biased opinion: you start with PyTorch because it's easy to develop and debug, and there's no point in having the fastest tools for a model that you can't train properly.

Once your model is running, and if/when you start hitting performance bottlenecks, then you consider migrating your model to TensorFlow.

[acgan]

Yup, this echoes the philosophy at e.g. Tesla. Development speed matters more than performance at first.

[albertzeyer]

But isn't TF eager mode just as easy to develop and debug, and the migration of TF eager mode to TF static mode is then probably simpler?

[reubenmorais]

TF eager mode has been a stable/supported thing for 10 days, since the release of 2.0. Before then it was available as opt-in behavior that once enabled meant all bets were off for if things would work or explode. So I think it's too early to answer your question. Maybe 2.0 bridges the gap to PyTorch in development speed. But maybe the momentum has already shifted to PyTorch.

[hprotagonist]

do you plan on having a research team?

[0-_-0]

How do you want to deploy your model?