Tensorflow 2.0 Beta 0
github.com
Tensorflow 2.0 Beta 0
1–10 of 44 posts
Re: Tensorflow 2.0 Beta 0
#2I am very happy that Google has realized the importance of usability. Hopefully that comes with concomitant improvements in the tf documentation, which, while thorough is completely unusable and lacks good examples for complex things.
Re: Tensorflow 2.0 Beta 0
#3Might 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.
Re: Tensorflow 2.0 Beta 0
#4Might 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.
Yeah, the only reason to use TF is really its deployment friendliness. If PyTorch addressed that more comprehensively, there'd be no good reason to use TF at all. For research PyTorch blows TF out of the water completely, and it's been that way for years, ever since it came out.
Re: Tensorflow 2.0 Beta 0
#5Might 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...
Re: Tensorflow 2.0 Beta 0
#6Might 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.
Yeah, the only reason to use TF is really its deployment friendliness. If PyTorch addressed that more comprehensively, there'd be no good reason to use TF at all. For research PyTorch blows TF out of the water completely, and it's been that way for years, ever since it came out.
What are you looking for in deployment friendliness? There's TorchScript to run your code faster (which is a work in progress)
Re: Tensorflow 2.0 Beta 0
#7Earlier quoted context omitted.
Yeah, the only reason to use TF is really its deployment friendliness. If PyTorch addressed that more comprehensively, there'd be no good reason to use TF at all. For research PyTorch blows TF out of the water completely, and it's been that way for years, ever since it came out.
What are you looking for in deployment friendliness? There's TorchScript to run your code faster (which is a work in progress)
One of the major benefits of TF 2.0 is apparently the capability to quickly deploy to TPU units with a single parameter change. (I haven't tried it, just followed the marketing).
AFAIK, This is still being worked on PyTorch via XLA, but not quite there yet.
Re: Tensorflow 2.0 Beta 0
#8Earlier quoted context omitted.
Yeah, the only reason to use TF is really its deployment friendliness. If PyTorch addressed that more comprehensively, there'd be no good reason to use TF at all. For research PyTorch blows TF out of the water completely, and it's been that way for years, ever since it came out.
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 a mobilenet-style model to e.g. ONNX or some other static graph format that does not require model code in order to load weights. I would like then to be able to load this saved model directly into something on Android and iOS that can use GPU and DSP present on the chip, with minimal extra futzing.
Re: Tensorflow 2.0 Beta 0
#9Might 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...
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.
Re: Tensorflow 2.0 Beta 0
#10Maybe 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 like my model is right in my hands. It's great for research and proofs-of-concept. Maybe for production too.