Link Search Menu Expand Document
enuSpace-tensorflow Manual

SpaceToBatch

tensorflow C++ API

tensorflow::ops::SpaceToBatch

SpaceToBatch for 4-D tensors of type T.

Summary

This is a legacy version of the more generalSpaceToBatchND.

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 theheightandwidthdimensions are moved to thebatchdimension. After the zero-padding, bothheightandwidthof 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 attrblock_sizemust 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:

SpaceToBatch block

Source link :https://github.com/EXPNUNI/enuSpaceTensorflow/blob/master/enuSpaceTensorflow/tf_array_ops.cpp

Argument:

  • Scope scope : A Scope object (A scope is generated automatically each page. A scope is not connected.)
  • Input input: 4-D with shape [batch, height, width, depth].
  • Input paddings: 2-D tensor of non-negative integers with shape [2, 2]
  • Int64 block_size: must be greater than one. It indicates the block size.

Output:

  • Output output: Output object of SpaceToBatch class object.

Result:

  • std::vector(Tensor) result_output: The output tensor.

Using Method


※ BatchToSpace와 반대되는 기능을 한다. input을 재정렬하며 input의 마지막 차원은 shape의 크기가 변하지 않는다.