undefined | Better HN

0 pointsNarew2y ago0 comments

Yes I understand why they do the move (they want to attract pytorch user). What's the benefit for the user instead of directly using pytorch for example ? I see we can maybe use tpu by switching to jax etc...

PS: sorry I'm a bit salty by my user experience of Keras.

0 comments

bootsmann2y ago

Pytorch is an animal by itself when you try to put it into production. They have started addressing it with torch 2.0 but it still has lengths to go. With this you can switch to TFserve if you have usual architecture.

cygn2y ago

You can just use Triton which is basically TFserve for Tensorflow, Pytorch, Onnx and more.

albertzeyer2y ago

Can you explain that?

My understand of Triton is more that this is an alternative to CUDA, but instead you write it directly in Python, and on a slightly higher-level, and it does a lot of optimizations automatically. So basically: Python -> Triton-IR -> LLVM-IR -> PTX.

https://openai.com/research/triton

1 more reply

m00x2y ago

Tensorflow has some advantages, like being able to use tf-lite for embedded devices. JAX is amazing on the TPU, which AFAIK pytorch doesn't have great support for.

I assume most people will still research in PyTorch, but then move it over the Keras for production models if they need multi-platform support.

minimaxir2y ago

Keras has a cleaner API compared to base PyTorch, especially if you want to use the Sequential construction as demoed in the post.

sva_2y ago

How so? You can use torch.nn.Sequential pretty much equivalently?

https://pytorch.org/docs/stable/generated/torch.nn.Sequentia...

minimaxir2y ago

Huh, didn't realize base PyTorch had an equivalent Sequential API.

The point about the better API overall still stands (notably including the actual training part, as base PyTorch requires you to implement your own loop)

2 more replies

j / k navigate · click thread line to collapse

0 comments

bootsmann2y ago

cygn2y ago

You can just use Triton which is basically TFserve for Tensorflow, Pytorch, Onnx and more.

albertzeyer2y ago

Can you explain that?

https://openai.com/research/triton

1 more reply

m00x2y ago

Tensorflow has some advantages, like being able to use tf-lite for embedded devices. JAX is amazing on the TPU, which AFAIK pytorch doesn't have great support for.

I assume most people will still research in PyTorch, but then move it over the Keras for production models if they need multi-platform support.

minimaxir2y ago

Keras has a cleaner API compared to base PyTorch, especially if you want to use the Sequential construction as demoed in the post.

sva_2y ago

How so? You can use torch.nn.Sequential pretty much equivalently?

https://pytorch.org/docs/stable/generated/torch.nn.Sequentia...

minimaxir2y ago

Huh, didn't realize base PyTorch had an equivalent Sequential API.

The point about the better API overall still stands (notably including the actual training part, as base PyTorch requires you to implement your own loop)

2 more replies

j / k navigate · click thread line to collapse