[npu]Add argsort op #34865
Merged
[npu]Add argsort op #34865
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
|
Thanks for your contribution! |
LielinJiang
previously approved these changes
Aug 13, 2021
qili93
reviewed
Aug 17, 2021
| limitations under the License. */ | ||
|
|
||
| #include <memory> | ||
| #include <string> |
| indices->mutable_data<int32_t>(ctx.GetPlace()); | ||
|
|
||
| int32_t axis = ctx.Attr<int>("axis"); | ||
| bool descending = ctx.Attr<bool>("descending"); |
There was a problem hiding this comment.
加一行 axis = (axis < 0) ? (in_dims.size() + axis) : axis; 使得axis永远为正数。
| framework::NPUAttributeMap sort_attr_input = {{"axis", npu_axis}, | ||
| {"descending", descending}}; | ||
|
|
||
| if (axis != -1) { |
There was a problem hiding this comment.
改为 if (axis == -1 || axis + 1 == in_dims.size()) {
| .stream(); | ||
| int32_t npu_axis = -1; | ||
| framework::NPUAttributeMap sort_attr_input = {{"axis", npu_axis}, | ||
| {"descending", descending}}; |
There was a problem hiding this comment.
冗余定义 npu_axis，直接写成
framework::NPUAttributeMap sort_attr_input = {{"axis", static_cast<int32_t>(-1)},
{"descending", descending}};
|
|
||
|
|
||
| @unittest.skipIf(not paddle.is_compiled_with_npu(), | ||
| "core is not compiled with NPU") |
There was a problem hiding this comment.
删除所有34-35行的类似代码
@unittest.skipIf(not paddle.is_compiled_with_npu(),
"core is not compiled with NPU")
LielinJiang
previously approved these changes
Aug 18, 2021
zhhsplendid
reviewed
Aug 19, 2021
| } | ||
| }; | ||
|
|
||
| template <typename T, typename Type> |
There was a problem hiding this comment.
What's the typename T used for?
| framework::Tensor* output) { | ||
| output->ShareDataWith(*input); | ||
| output->Resize(framework::make_ddim(std::move(input_shapes))); | ||
| input = output; |
There was a problem hiding this comment.
What's the purpose of this statement?
| .AddInput(input_reshape_tensor) | ||
| .AddOutput(input_scatter_tensor); | ||
| runner_scatter.Run(stream); | ||
| ReshapeNPU<T, Type>(ctx, &input_scatter_tensor, trans_shapes, t_out); |
There was a problem hiding this comment.
I saw t_out is ShareDataWith and a t_out can be dX, which is the output of the ArgsortGradNPUKernel, this is not allowed in PaddlePaddle.
zhhsplendid
reviewed
Aug 19, 2021
| }; | ||
|
|
||
| template <typename Type> | ||
| static void ReshapeNPU(const framework::ExecutionContext& ctx, |
qili93
approved these changes
Aug 20, 2021
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
New features
PR changes
OPs
Describe
[NPU] Support npu kernel for argsort and argsort_grad
