Class MultiDeviceIterator

java.lang.Object
org.tensorflow.op.RawOp
org.tensorflow.op.data.MultiDeviceIterator
All Implemented Interfaces:
Shaped, Op, Operand<TType>

@Operator(group="data") public final class MultiDeviceIterator extends RawOp implements Operand<TType>
Creates a MultiDeviceIterator resource.
  • Field Details

  • Constructor Details

    • MultiDeviceIterator

      public MultiDeviceIterator(Operation operation)
  • Method Details

    • create

      @Endpoint(describeByClass=true) public static MultiDeviceIterator create(Scope scope, List<String> devices, String sharedName, String container, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes)
      Factory method to create a class wrapping a new MultiDeviceIterator operation.
      Parameters:
      scope - current scope
      devices - A list of devices the iterator works across.
      sharedName - If non-empty, this resource will be shared under the given name across multiple sessions.
      container - If non-empty, this resource is placed in the given container. Otherwise, a default container is used.
      outputTypes - The type list for the return values.
      outputShapes - The list of shapes being produced.
      Returns:
      a new instance of MultiDeviceIterator
    • handle

      public Output<? extends TType> handle()
      Gets handle. Handle to the resource created.
      Returns:
      handle.
    • asOutput

      public Output<TType> asOutput()
      Description copied from interface: Operand
      Returns the symbolic handle of the 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.

      Specified by:
      asOutput in interface Operand<TType>
      See Also: