Hacker News Reader

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.

View on HN

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

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?

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.

Reply on news.ycombinator.com