An op that groups a list of partitioned inputs together. Supports ND sharding.
Nested Classes
class | TPUPartitionedInputV2.Options | Optional attributes for TPUPartitionedInputV2
|
Public Methods
Output<T> |
asOutput()
Returns the symbolic handle of a tensor.
|
static <T> TPUPartitionedInputV2<T> |
create(Scope scope, Iterable<Operand<T>> inputs, List<Long> partitionDims, Options... options)
Factory method to create a class wrapping a new TPUPartitionedInputV2 operation.
|
static TPUPartitionedInputV2.Options |
isPacked(Boolean isPacked)
|
Output<T> |
output()
A handle which represents the full shape of partitioned tensors.
|
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 TPUPartitionedInputV2<T> create (Scope scope, Iterable<Operand<T>> inputs, List<Long> partitionDims, Options... options)
Factory method to create a class wrapping a new TPUPartitionedInputV2 operation.
Parameters
scope | current scope |
---|---|
inputs | A list of partitioned inputs which must have the same shape. |
partitionDims | A list of integers describing how each dimension is partitioned. Emptiness indicates the inputs are replicated. |
options | carries optional attributes values |
Returns
- a new instance of TPUPartitionedInputV2
public static TPUPartitionedInputV2.Options isPacked (Boolean isPacked)
Parameters
isPacked | Indicates whether the input is a packed resource. |
---|