AssignVariableXlaConcatND

public final class AssignVariableXlaConcatND

Concats input tensor across all dimensions.

An op which merges slices the input tensor based on the given num_splits attribute, strips paddings optionally, and writes the merged tensor without paddings to the resource variable.

This op may be generated via the TPU bridge.

For example, with `input` tensor:

[[0, 1],
  [4, 5]]
 [[2, 3],
  [6, 7]]
 [[8, 9],
  [12, 13]]
 [[10, 11],
  [14, 15]]
 
`num_splits`:
[2, 2]
 
and `paddings`:
[1, 1]
 
the expected `outputs` is:
[[0, 1, 2],
  [4, 5, 6],
  [8, 9, 10]]
 

Nested Classes

class AssignVariableXlaConcatND.Options Optional attributes for AssignVariableXlaConcatND  

Public Methods

static <T> AssignVariableXlaConcatND
create(Scope scope, Operand<?> resource, Iterable<Operand<T>> inputs, List<Long> numConcats, Options... options)
Factory method to create a class wrapping a new AssignVariableXlaConcatND operation.
static AssignVariableXlaConcatND.Options
paddings(List<Long> paddings)

Inherited Methods

Public Methods

public static AssignVariableXlaConcatND create (Scope scope, Operand<?> resource, Iterable<Operand<T>> inputs, List<Long> numConcats, Options... options)

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

Parameters
scope current scope
resource Resource variable for concatenated input tensors across all dimensions. } in_arg { name: "inputs" description: <
numConcats Number of ways to merge per dimension.
options carries optional attributes values
Returns
  • a new instance of AssignVariableXlaConcatND

public static AssignVariableXlaConcatND.Options paddings (List<Long> paddings)

Parameters
paddings Optional list of right paddings per dimension to strip from the final merged tensor. These paddings must not exceed the dimension size of the merged result prior to stripping paddings.