How to convert this vectorized code into GPU code for MAXIMUM speedup ?

2017 4 13
2 个回答
更新时间：2017 4 26
9 次查看（30 天）

回答（2 个）

Joss Knight
Joss Knight 2017-4-18

0 个投票

I was able to get a marginal speedup with additional vectorization of the mask:
x = sum(I < cat(3, 120, 155, 160), 3) == true;
but otherwise you've done pretty well. You've got to wonder why you need to replicate the output on every channel however. Why not discard the colour channels if you're using grayscale?

2 个评论
显示 无 隐藏 无

Tanmay Virnodkar
Tanmay Virnodkar 2017-4-20
编辑：Tanmay Virnodkar 2017-4-20
Actually i want to show speedup or lets say difference between normal cpu time and gpu time..... so if i first convert the picture into grayscale and use imtool then cpu time is also very less so i am not able to show speed up ...hence i decided not to discard R G B channels.
Joss Knight
Joss Knight 2017-4-26
Right, but then you're including the cost of replicating data in GPU memory and doing indexing, which is memory-bound and doesn't necessarily show the GPU in a great light.

请先登录，再进行评论。

Jan
Jan 2017-4-18

0 个投票

The bottlenecks of the code are the darn clear all and the disk access using imwrite. Moving this to the GPU will not help.

类别

在 帮助中心 和 File Exchange 中查找有关 GPU Computing 的更多信息

标签

评论：

Jan
Jan
2017-4-26

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by