TPUReplicatedInput

public final class TPUReplicatedInput

Connects N inputs to an N-way replicated TPU computation.

This operation holds a replicated input to a `tpu.replicate()` computation subgraph. Each replicated input has the same shape and type alongside the output.

For example:

%a = "tf.opA"()
 %b = "tf.opB"()
 %replicated_input = "tf.TPUReplicatedInput"(%a, %b)
 %computation = "tf.Computation"(%replicated_input)
 
The above computation has a replicated input of two replicas.

Nested Classes

class TPUReplicatedInput.Options Optional attributes for TPUReplicatedInput  

Public Methods

Output<T>
asOutput()
Returns the symbolic handle of a tensor.
static <T> TPUReplicatedInput<T>
create(Scope scope, Iterable<Operand<T>> inputs, Options... options)
Factory method to create a class wrapping a new TPUReplicatedInput operation.
static TPUReplicatedInput.Options
index(Long index)
static TPUReplicatedInput.Options
isMirroredVariable(Boolean isMirroredVariable)
static TPUReplicatedInput.Options
isPacked(Boolean isPacked)
Output<T>
output()

Inherited Methods

Public Methods

public Output<T> 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 TPUReplicatedInput<T> create (Scope scope, Iterable<Operand<T>> inputs, Options... options)

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

Parameters
scope current scope
options carries optional attributes values
Returns
  • a new instance of TPUReplicatedInput

public static TPUReplicatedInput.Options index (Long index)

public static TPUReplicatedInput.Options isMirroredVariable (Boolean isMirroredVariable)

public static TPUReplicatedInput.Options isPacked (Boolean isPacked)

public Output<T> output ()