Apply to speak at TensorFlow World. Deadline April 23rd. Propose talk

tensorflow::ops::ExtractVolumePatches

#include <array_ops.h>

Extract patches from input and put them in the "depth" output dimension.

Summary

3D extension of extract_image_patches.

Arguments:

  • scope: A Scope object
  • input: 5-D Tensor with shape [batch, in_planes, in_rows, in_cols, depth].
  • ksizes: The size of the sliding window for each dimension of input.
  • strides: 1-D of length 5. How far the centers of two consecutive patches are in input. Must be: [1, stride_planes, stride_rows, stride_cols, 1].
  • padding: The type of padding algorithm to use.

We specify the size-related attributes as:

      ksizes = [1, ksize_planes, ksize_rows, ksize_cols, 1]
      strides = [1, stride_planes, strides_rows, strides_cols, 1]

Returns:

  • Output: 5-D Tensor with shape [batch, out_planes, out_rows, out_cols, ksize_planes * ksize_rows * ksize_cols * depth] containing patches with size ksize_planes x ksize_rows x ksize_cols x depth vectorized in the "depth" dimension. Note out_planes, out_rows and out_cols are the dimensions of the output patches.

Constructors and Destructors

ExtractVolumePatches(const ::tensorflow::Scope & scope, ::tensorflow::Input input, const gtl::ArraySlice< int > & ksizes, const gtl::ArraySlice< int > & strides, StringPiece padding)

Public attributes

operation
patches

Public functions

node() const
::tensorflow::Node *
operator::tensorflow::Input() const
operator::tensorflow::Output() const

Public attributes

operation

Operation operation

patches

::tensorflow::Output patches

Public functions

ExtractVolumePatches

 ExtractVolumePatches(
  const ::tensorflow::Scope & scope,
  ::tensorflow::Input input,
  const gtl::ArraySlice< int > & ksizes,
  const gtl::ArraySlice< int > & strides,
  StringPiece padding
)

node

::tensorflow::Node * node() const 

operator::tensorflow::Input

 operator::tensorflow::Input() const 

operator::tensorflow::Output

 operator::tensorflow::Output() const