[PHI]Move sum op to PHI #45860
Merged
[PHI]Move sum op to PHI #45860
Conversation
This file contains hidden or 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! |
zyfncg
reviewed
Sep 13, 2022
Comment on lines
27
to
33
| if (x.size() > 0 && x[0]->initialized() && DenseTensor::classof(x[0])) { | ||
| if ((static_cast<const DenseTensor*>(x[0]))->Holder() == out->Holder()) { | ||
| in_place = true; | ||
| } | ||
| } | ||
|
|
||
| if (in_num >= 1 && DenseTensor::classof(x[0]) && x[0]->initialized()) { |
| auto in = EigenVector<T>::Flatten(in_t); | ||
| result.device(place) = result + in; | ||
| } | ||
| VLOG(10) << "end sum kernel"; |
|
|
||
| template <typename T, typename Context> | ||
| void AddNKernel(const Context& dev_ctx, | ||
| const std::vector<const SelectedRows*>& x, |
Contributor
There was a problem hiding this comment.
有了 std::vector<const TensorBase*> 的kernel还需要std::vector<const SelectedRows*>类型的吗？
chenwhql
reviewed
Sep 13, 2022
Contributor
chenwhql
left a comment
chenwhql
left a comment
There was a problem hiding this comment.
还有 add_n的python api下面也需要去掉is_dense的判断分支
| namespace phi { | ||
|
|
||
| template <typename T, typename Context> | ||
| void AddNKernel(const Context& dev_ctx, |
Contributor
There was a problem hiding this comment.
加注释说明一下，原则上TensorBase不允许作为kernel输入类型，这里主要是为了兼容fluid
|
|
||
| ////////////////// Forward api impls ////////////////////// | ||
|
|
||
| Tensor add_n_impl(const std::vector<Tensor>& x) { |
Contributor
Author
There was a problem hiding this comment.
已添加选择add_n_sr的逻辑，3q
Contributor
Author
done |
zyfncg
approved these changes
Sep 16, 2022
zkh2016
pushed a commit
to zkh2016/Paddle
that referenced
this pull request
Oct 13, 2022
This was referenced Dec 26, 2022
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
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
Describe
该pr主要做了如下工作：
1，将sum op的kernel迁移至phi并重构InferMeta进行兼容适配
2，混合Tensor类型的vector输入的算子迁移机制开发完善
3，InferShape兼容组件在SelectedRows情景下的多处bug修复
4，MetaTensor功能扩展，支持Tensor类型判断。