TensorFlow 1.11 released | Hacker News Reader

TensorFlow 1.11 released

github.com

55 pointsjamesblonde10 comments

The main highlight is a new version of AllReduce native to TensorFlow. This, like Horovod, builds a ring and has both native and NCCL implementations. Early results (albeit on a single host) show that it outperforms Horovod: https://groups.google.com/a/tensorflow.org/forum/#!topic/dis...

And it looks like it's natively supported via Keras too? https://github.com/tensorflow/tensorflow/tree/master/tensorf...

Yes, and with the estimator API, too.

> Add multi-GPU DistributionStrategy support in tf.keras. Users can now use fit, evaluate and predict to distribute their model on multiple GPUs.

Is this different than the existing multi_gpu_model? (https://keras.io/utils/#multi_gpu_model)

Looks like Python 3.7 is not supported yet:

https://github.com/tensorflow/tensorflow/issues/20517

Somewhat off topic: I'm rediscovering the dance required to set up a new Windows 10 laptop with Python, CUDA, and TensorFlow. If I understand correctly, pyenv does not work on Windows. Is there something similar?

Also, if I installed TensorFlow in WSL, it wouldn't be able to use the GPU, would it?

Python 3.7 is supported if you build master from source. But that can be a lot of work and though it may solve your TF problem, it may create other problems with other Python packages.

In my experience conda is the way to go for setting up your environment. I did not have great luck with pyenv, YMMV.

I don't know about WSL. Maybe setup a dual boot with Ubuntu on your system?

No, CUDA is not supported by WSL. They hope they can support it at some point, though. Use Conda on Linux.

You can use cygwin, which does work with the GPU. Not sure if that works with pyenv but it does work with pipenv...

I am using conda environments.

Does it still have that horrible java build dependency?

Reply on news.ycombinator.com