tensorflow::ops::SpaceToBatch

tensorflow::ops::SpaceToBatch

#include <array_ops.h>

SpaceToBatch for 4-D tensors of type T.

Summary

This is a legacy version of the more general SpaceToBatchND.

Zero-pads and then rearranges (permutes) blocks of spatial data into batch. More specifically, this op outputs a copy of the input tensor where values from the height and width dimensions are moved to the batch dimension. After the zero-padding, both height and width of the input must be divisible by the block size.

Arguments:

  • scope: A Scope object
  • input: 4-D with shape [batch, height, width, depth].
  • paddings: 2-D tensor of non-negative integers with shape [2, 2]. It specifies the padding of the input with zeros across the spatial dimensions as follows:
    paddings = [[pad_top, pad_bottom], [pad_left, pad_right]]
    
    The effective spatial dimensions of the zero-padded input tensor will be:
    height_pad = pad_top + height + pad_bottom
    width_pad = pad_left + width + pad_right
    

The attr block_size must be greater than one. It indicates the block size.

  • Non-overlapping blocks of size block_size x block size in the height and width dimensions are rearranged into the batch dimension at each location.
  • The batch of the output tensor is batch * block_size * block_size.
  • Both height_pad and width_pad must be divisible by block_size.

The shape of the output will be:

[batch*block_size*block_size, height_pad/block_size, width_pad/block_size,
 depth]

Some examples:

(1) For the following input of shape [1, 2, 2, 1] and block_size of 2:

``` x = [[[[1], [2]], [[3], [4]]]] ```

The output tensor has shape [4, 1, 1, 1] and value:

``` [[[[1]]], [[[2]]], [[[3]]], [[[4]]]] ```

(2) For the following input of shape [1, 2, 2, 3] and block_size of 2:

``` x = [[[[1, 2, 3], [4, 5, 6]], [[7, 8, 9], [10, 11, 12]]]] ```

The output tensor has shape [4, 1, 1, 3] and value:

``` [[[1, 2, 3]], [[4, 5, 6]], [[7, 8, 9]], [[10, 11, 12]]] ```

(3) For the following input of shape [1, 4, 4, 1] and block_size of 2:

``` x = [[[[1], [2], [3], [4]], [[5], [6], [7], [8]], [[9], [10], [11], [12]], [[13], [14], [15], [16]]]] ```

The output tensor has shape [4, 2, 2, 1] and value:

``` x = [[[[1], [3]], [[9], [11]]], [[[2], [4]], [[10], [12]]], [[[5], [7]], [[13], [15]]], [[[6], [8]], [[14], [16]]]] ```

(4) For the following input of shape [2, 2, 4, 1] and block_size of 2:

``` x = [[[[1], [2], [3], [4]], [[5], [6], [7], [8]]], [[[9], [10], [11], [12]], [[13], [14], [15], [16]]]] ```

The output tensor has shape [8, 1, 2, 1] and value:

``` x = [[[[1], [3]]], [[[9], [11]]], [[[2], [4]]], [[[10], [12]]], [[[5], [7]]], [[[13], [15]]], [[[6], [8]]], [[[14], [16]]]] ```

Among others, this operation is useful for reducing atrous convolution into regular convolution.

Returns:

Constructors and Destructors
SpaceToBatch(const ::tensorflow::Scope & scope, ::tensorflow::Input input, ::tensorflow::Input paddings, int64 block_size)
Public attributes
output
Public functions
node() const
::tensorflow::Node *
operator::tensorflow::Input() const
operator::tensorflow::Output() const

Public attributes

output

::tensorflow::Output output

Public functions

SpaceToBatch

 SpaceToBatch(
  const ::tensorflow::Scope & scope,
  ::tensorflow::Input input,
  ::tensorflow::Input paddings,
  int64 block_size
)

node

::tensorflow::Node * node() const 

operator::tensorflow::Input

operator::tensorflow::Input() const 

operator::tensorflow::Output

operator::tensorflow::Output() const 

© 2017 The TensorFlow Authors. All rights reserved.
Licensed under the Creative Commons Attribution License 3.0.
Code samples licensed under the Apache 2.0 License.
https://www.tensorflow.org/api_docs/cc/class/tensorflow/ops/space-to-batch.html

在线笔记
App下载
App下载

扫描二维码

下载编程狮App

公众号
微信公众号

编程狮公众号

意见反馈
返回顶部