News for hackers. Stuff that matters.

A comment on Keras 3.0. Parent comment

Re:Keras 3.0 by pennaMan on Tuesday November 28, 2023 @10:41AM (#38446871)
What is relevant now?
[ Parent | on HN ]
Re:Keras 3.0 by elforce002 on Tuesday November 28, 2023 @11:47AM (#38447765)
I use keras for CV but for DL at work, we started moving towards pytorch and closely following modular.
[ Parent | on HN ]
Re:Keras 3.0 by arketyp on Tuesday November 28, 2023 @10:48AM (#38446969)
PyTorch has the nicest debugging experience. JAX is supposed to be powerful too, but I'm skeptical of its static graph which is what made TensorFlow such a chore. I'm whishful about Mojo, in theory it sounds great.
[ Parent | on HN ]
Re:Keras 3.0 by techwizrd on Tuesday November 28, 2023 @11:07AM (#38447234)
On the contrary, that's why I find Keras 3 interesting (although I don't see myself switching anytime soon). It supports both PyTorch and JAX, and Keras Core models/layers can be used interchangeably with PyTorch modules: https://twitter.com/fchollet/status/1697020631752298982
[ Parent | on HN ]