[Operator] optimize normalize op with vectorized load, dynamic shape and more #316
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
|
the tests are passing |
| @@ -104,39 +108,69 @@ def allow_prologue(self) -> bool: | |||
| return False | |||
|
|
|||
| def allow_epilogue(self) -> bool: | |||
There was a problem hiding this comment.
问一个比较小白的问题，prologue 和这里的 epilogue，在中文里面一般怎么翻译，二者是什么作用呢，谢谢
There was a problem hiding this comment.
Please refer to the section 4.2 and section 5.2 of our paper[1] to learn more about about the prologue and epilogue fusion (in the paper, it's called post-scheduling-fusion). It seems that there is no obvious translation in Chinese, maybe "前驱算子" and "后继算子".
vadiklyutiy
pushed a commit
that referenced
this pull request
Jul 22, 2024
Previously I added hidet.ones_like without handling dtype and device. Now made it properly work with all these arguments --------- Co-authored-by: Zhumakhan <nazirzhumakhan@gmail,.com>
vadiklyutiy
pushed a commit
that referenced
this pull request
Jul 23, 2024
Previously I added hidet.ones_like without handling dtype and device. Now made it properly work with all these arguments --------- Co-authored-by: Zhumakhan <nazirzhumakhan@gmail,.com>
vadiklyutiy
pushed a commit
that referenced
this pull request
Dec 26, 2024
Previously I added hidet.ones_like without handling dtype and device. Now made it properly work with all these arguments --------- Co-authored-by: Zhumakhan <nazirzhumakhan@gmail,.com>
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Outdated suggestions cannot be applied.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.
this change introduces several enhancements to the current norm operator
as a result, norm_fp16.py can be safely deleted
for shapes in stable diffusion: [2, 32, 60, 16, 16], norm dims [60, 16, 16], fp32:
for shapes in bert-base: [1, 128, 768], norm dims [768], fp16:
We are still not faster than torch, but very close