Is Tensorflow Federated-Learning only for simulating federated learning on one machine?

问题

I read multiple guides on https://www.tensorflow.org/federated/federated_learning e.g. the image classification or text generation example.

From what I have read I can not see how to use tensorflow federated-learning (tff) for a real world application: datasets on multiple hardware clients. It all looks like its meant only for simulating federated learning.

I want to use tff on multiple machines and not simulate it on only one. I would appreciate it when someone knows if it's even possible with tff or found a guide on how to do it.

thank you.

回答1:

As of today TFF only provides a simulation environment for use in Federated Learning (FL) research.

There is work being done on a supporting multi-machine simulation environment, but this is still ongoing work (see https://github.com/tensorflow/federated/blob/master/tensorflow_federated/python/core/impl/remote_executor.py)

There is not yet a "real world" FL deployment platform.

来源：https://stackoverflow.com/questions/57323598/is-tensorflow-federated-learning-only-for-simulating-federated-learning-on-one-m