TensorFlow 1 version View source on GitHub

Simple implementation of ClusterResolver that accepts a ClusterSpec.

Inherits From: ClusterResolver

environment Returns the current environment which TensorFlow is running in.

There are two possible return values, "google" (when TensorFlow is running in a Google-internal environment) or an empty string (when TensorFlow is running elsewhere).

If you are implementing a ClusterResolver that works in both the Google environment and the open-source world (for instance, a TPU ClusterResolver or similar), you will have to return the appropriate string depending on the environment, which you will have to detect.

Otherwise, if you are implementing a ClusterResolver that will only work in open-source TensorFlow, you do not need to implement this property.






View source

Returns the ClusterSpec passed into the constructor.


View source

Returns the master address to use when creating a session.

task_type (Optional) The type of the TensorFlow task of the master.
task_id (Optional) The index of the TensorFlow task of the master.
rpc_layer (Optional) The RPC used by distributed TensorFlow.

The name or URL of the session master.

If a task_type and task_id is given, this will override the master string passed into the initialization function.


View source

Returns the number of accelerator cores per worker.

The SimpleClusterResolver does not do automatic detection of accelerators, so a TensorFlow session will never be created, and thus all arguments are unused and we simply assume that the type of accelerator is a GPU and return the value in provided to us in the constructor.

task_type Unused.
task_id Unused.
config_proto Unused.