【Hackathon No.33】为 Paddle 优化 erfinv op 在 GPU 上的计算性能 #199
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
JamesLim-sy
reviewed
Aug 11, 2022
| | Case No. | device | input_shape | input_type | Paddle Perf(ms) | | ||
| |---|---|---|---|---| | ||
| | 1 | RTX 2070s | [-1L, 204800L] | float32 | 0.1438 | | ||
| | 2 | RTX 2070s |[10L, 20L, 30L, 40L, 5L, 6L] | float64 8| 8.6485 | |
|
|
||
| Pytorch中对Erfinv算子的实现基于GPU计算, forward整体性能如下(基于pytorch v1.12)： | ||
|
|
||
| | Case No. | device | input_shape | input_type | Paddle Perf(ms) | |
There was a problem hiding this comment.
Paddle Perf(ms) 这部分是不是应该改成 Pytorch Perf(ms)
|
|
||
| ## 2.1 关键模块与性能提升点 | ||
|
|
||
| 通过使用飞桨内部的Elementwise Kernel来进行计算。通过向量化读取、向量化写入以及gpu_launch_config.h中的线程配置方法对算子进行优化，预计提升1.2倍。 |
There was a problem hiding this comment.
性能提升预估1.2x倍提升后，数值上之后距离torch的性能还有差异，可以尝试看下底层C++端二者是否还有什么实现差异。
There was a problem hiding this comment.
@JamesLim-sy 尝试了torch的c++实现方式，也尝试了ndtri函数实现，性能没有明显提升。最终使用cuda内置函数，得到了２倍以上的提升，相比torch也有１倍以上的提升。
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
为 Paddle 优化 erfinv op 在 GPU 上的计算性能
任务：PaddlePaddle/Paddle#44072 (comment)