[PIR] Improve Dtype Transfer (add infermeta by value) #60677
Merged
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
|
你的PR提交成功，感谢你对开源项目的贡献! |
|
请完善一下 pr 描述 |
chen2016013
changed the title
zhangbo9674
approved these changes
Jan 15, 2024
| return get_attributes_str | ||
|
|
||
|
|
||
| def gen_infermeta_func_str( |
There was a problem hiding this comment.
op_build_gen.py 文件中原本算子的 Build 函数中，可以替换为该infer_meta 接口？
| @@ -177,6 +177,71 @@ void AddNOp::InferMeta(phi::InferMetaContext *infer_meta) { | |||
| fn(infer_meta); | |||
| } | |||
|
|
|||
| std::vector<pir::Type> AddNOp::InferMeta( | |||
| "DenseTensorType and SelectedRowsType.")); | ||
| } | ||
| } | ||
|
|
||
| std::string GetValueDataType(const pir::Value& value) { |
There was a problem hiding this comment.
可以调用新增的GetValueDataType(pir::Type type)
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Outdated suggestions cannot be applied.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.
PR types
Others
PR changes
Others
Description
背景： 在运算中，有一种场景如下：对于某个二元运算op，输入x.dype()=float64，y.dtype()=complex128，infermeta结果float64（按照第一个输入确定）。在pass中会对x做data transfer以便和y进行计算，此时x.dype()=y.dtype()=complex128，但op_result的结果及后续使用该输出的相关op的输入输出类型没有做transfer，仍然是float64，导致错误。
修改方案：在构建kernel op output前，对于所有op再进行一轮Infermeta。具体地，将Infermeta抽象出相应接口在pd_op_to_kernel_pass中调用
特例：用户上层组网时可能指定输出的 dtype，这部分dtype不应该被改变
Pcard-67164