tensorflow::ops::TopK

#include <nn_ops.h>

Finds values and indices of the k largest elements for the last dimension.

Summary

If the input is a vector (rank-1), finds the k largest entries in the vector and outputs their values and indices as vectors. Thus values[j] is the j-th largest entry in input, and its index is indices[j].

For matrices (resp. higher rank input), computes the top k entries in each row (resp. vector along the last dimension). Thus,

values.shape = indices.shape = input.shape[:-1] + [k]

If two elements are equal, the lower-index element appears first.

Args:

  • scope: A Scope object
  • input: 1-D or higher with last dimension at least k.
  • k: 0-D. Number of top elements to look for along the last dimension (along each row for matrices).

Optional attributes (see Attrs):

  • sorted: If true the resulting k elements will be sorted by the values in descending order.

Returns:

  • Output values: The k largest elements along each last dimensional slice.
  • Output indices: The indices of values within the last dimension of input.

Constructors and Destructors

TopK(const ::tensorflow::Scope & scope, ::tensorflow::Input input, ::tensorflow::Input k)
TopK(const ::tensorflow::Scope & scope, ::tensorflow::Input input, ::tensorflow::Input k, const TopK::Attrs & attrs)

Public attributes

indices
operation
values

Public static functions

IndexType(DataType x)
Sorted(bool x)

Structs

tensorflow::ops::TopK::Attrs

Optional attribute setters for TopK.

Public attributes

indices

::tensorflow::Output indices

operation

Operation operation

values

::tensorflow::Output values

Public functions

TopK

 TopK(
  const ::tensorflow::Scope & scope,
  ::tensorflow::Input input,
  ::tensorflow::Input k
)

TopK

 TopK(
  const ::tensorflow::Scope & scope,
  ::tensorflow::Input input,
  ::tensorflow::Input k,
  const TopK::Attrs & attrs
)

Public static functions

IndexType

Attrs IndexType(
  DataType x
)

Sorted

Attrs Sorted(
  bool x
)