CUDALink`
CUDALink`

CUDATranspose

CUDATranspose[mat]

transposes input matrix mat.

CUDATranspose[mem]

transposes input CUDAMemory mem.

更多信息和选项

  • The CUDALink application must be loaded using Needs["CUDALink`"].
  • CUDATranspose works on matrices.
  • CUDATranspose works with binary or grayscale images.
  • CUDATranspose works with CUDAMemory, but cannot work on fixed vector struct types like "Float[2]", "Integer32[2]", ….
  • CUDATranspose does not work on symbolic input.
  • CUDATranspose is equivalent to Transpose.

范例

打开所有单元关闭所有单元

基本范例  (2)

First, load the CUDALink application:

This transposes an input matrix:

If input is CUDAMemory, then memory is returned as output:

The transposes the input CUDAMemory:

Memory is retrieved using CUDAMemoryGet:

Scope  (2)

Input can be an image, but must be single-channeled:

Image components can be split and transposed independently:

The resulting color-separated image can be combined:

Wolfram Research (2010)，CUDATranspose，Wolfram 语言函数，https://reference.wolfram.com/language/CUDALink/ref/CUDATranspose.html.

文本

Wolfram Research (2010)，CUDATranspose，Wolfram 语言函数，https://reference.wolfram.com/language/CUDALink/ref/CUDATranspose.html.

CMS

Wolfram 语言. 2010. "CUDATranspose." Wolfram 语言与系统参考资料中心. Wolfram Research. https://reference.wolfram.com/language/CUDALink/ref/CUDATranspose.html.

APA

Wolfram 语言. (2010). CUDATranspose. Wolfram 语言与系统参考资料中心. 追溯自 https://reference.wolfram.com/language/CUDALink/ref/CUDATranspose.html 年

BibTeX

@misc{reference.wolfram_2024_cudatranspose, author="Wolfram Research", title="{CUDATranspose}", year="2010", howpublished="\url{https://reference.wolfram.com/language/CUDALink/ref/CUDATranspose.html}", note=[Accessed: 21-November-2024 ]}

BibLaTeX

@online{reference.wolfram_2024_cudatranspose, organization={Wolfram Research}, title={CUDATranspose}, year={2010}, url={https://reference.wolfram.com/language/CUDALink/ref/CUDATranspose.html}, note=[Accessed: 21-November-2024 ]}