How to run TF learn (skflow) Kmeans clustering in multi-machine multi-gpu environment?

Question

Is there any example / documentation which shows how to implement skflow(tf.contrib.learn) in multi-machine and multi-GPUs environment?

I want to implement KMeansClustring using skflow(tf.contrib.learn) in a multi-machine and multi-GPUs environment. How should I proceed?

Answer 1

Ideally you'd be running a cluster manager (Kubernetes, Mesos), otherwise setting up multi-machine training is rather difficult.

For a core TensorFlow example, see here (under the docker/kubernetes directories): https://github.com/tensorflow/ecosystem

We'll update the examples there to use both Estimator and multi-GPU setups soon.