Warning: This API is deprecated and will be removed in a future version of TensorFlow after the replacement is stable.

ConfigureTPUEmbeddingHost

Stay organized with collections Save and categorize content based on your preferences.
public final class ConfigureTPUEmbeddingHost

An op that configures the TPUEmbedding software on a host.

Public Methods

Output<String>
asOutput()
Returns the symbolic handle of a tensor.
static ConfigureTPUEmbeddingHost
create(Scope scope, Operand<String> commonConfig, Operand<String> memoryConfig, String config)
Factory method to create a class wrapping a new ConfigureTPUEmbeddingHost operation.
Output<String>
networkConfig()
A string containing metadata about the hostname and RPC port used for communication with this host.

Inherited Methods

Public Methods

public Output<String> asOutput ()

Returns the symbolic handle of a tensor.

Inputs to TensorFlow operations are outputs of another TensorFlow operation. This method is used to obtain a symbolic handle that represents the computation of the input.

public static ConfigureTPUEmbeddingHost create (Scope scope, Operand<String> commonConfig, Operand<String> memoryConfig, String config)

Factory method to create a class wrapping a new ConfigureTPUEmbeddingHost operation.

Parameters
scope current scope
commonConfig A string-encoded common configuration proto containing metadata about the TPUEmbedding partitioner output.
memoryConfig A string-encoded memory config proto containing metadata about the memory allocations reserved for TPUEmbedding.
config An TPUEmbeddingConfiguration proto serialized to a string, describing the desired TPUEmbedding configuration.
Returns
  • a new instance of ConfigureTPUEmbeddingHost

public Output<String> networkConfig ()

A string containing metadata about the hostname and RPC port used for communication with this host.