Tensorflow 2.0 Beta 0

Post by chadmeister » Fri, Jun 07, 2019, 8:11 PM UTC

m0zg wrote: post #8

forgotmyhnacc wrote: post #6

Earlier quoted context omitted.

What are you looking for in deployment friendliness? There's TorchScript to run your code faster (which is a work in progress)

For me it's deploying to mobile, mostly. There's ONNX but it doesn't seem to be terribly mature and it doesn't support some of the common ops, and e.g. FB's own Caffe2 doesn't run it natively. There's also no mature tooling to produce quantized models. TF remains the only real option to do quantization aware training or even easy post-training quantization. Specifically, my life would be a lot easier if I could save…

Post by p1esk » Fri, Jun 07, 2019, 8:23 PM UTC

cheez wrote: post #11

sytelus wrote: post #9

Earlier quoted context omitted.

This is one thing that confuses me. Why Keras is still a separate brand? Why everything isn't under just tensorflow namespace instead of having to do tf.keras all the time. I really wish tf just had one API and just one thing to learn.

Keras is a high level API that can use multiple backends. So it makes sense for them to remain separated.

Post by levesque » Fri, Jun 07, 2019, 9:12 PM UTC

sandGorgon wrote: post #5

levesque wrote: post #3

Might give it another try, but my latest incursion in the Tensorflow universe did not end pleasantly. I ended up recoding everything in Pytorch, took me less than a day to do the stuff that took me more than a week in TF. One problem is that there are too many ways to do the same thing in TF and it's hard to transition from one to the other.

What do you think of Keras in this space ? Because TF 2.0 is entirely keras based. https://medium.com/tensorflow/standardizing-on-keras-guidanc...

Post by strebler » Fri, Jun 07, 2019, 9:40 PM UTC

NickHoff wrote: post #10

Maybe I'll give TF another try, but right now I'm really liking PyTorch. With TensorFlow I always felt like my models were buried deep in the machine and it was very hard to inspect and change them, and if I wanted to do something non-standard (which for me is most of the time) it was difficult even with Keras. With PyTorch though, I connect things however how I want, write whatever training logic I want, and I feel…

Post by panpanna » Fri, Jun 07, 2019, 9:45 PM UTC

strebler wrote: post #18

NickHoff wrote: post #10

Maybe I'll give TF another try, but right now I'm really liking PyTorch. With TensorFlow I always felt like my models were buried deep in the machine and it was very hard to inspect and change them, and if I wanted to do something non-standard (which for me is most of the time) it was difficult even with Keras. With PyTorch though, I connect things however how I want, write whatever training logic I want, and I feel…

TF's deprecation velocity was way too high for my taste. Things we wrote would stop working randomly with their updates. I feel very similar to you about the models being "buried too deep" in their (ever-changing) machine. I much preferred how easy it was to hack Caffe V1 (once you got past the funky names, etc). These days, I really like mxnet. Torch was a disaster, but Pytorch is much better. It's not bad in produc…