Move one hot to phi #39876
Merged
Move one hot to phi #39876
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! |
… move_one_hot_to_phi
… move_one_hot_to_phi
zyfncg
reviewed
Feb 25, 2022
| auto out_dims_vec = phi::vectorize(x_dims); | ||
| out_dims_vec.push_back(depth); | ||
| auto out_dims = phi::make_ddim(out_dims_vec); | ||
| out->set_dims(out_dims); |
| } | ||
| }; | ||
| out->mutable_data<T>(dev_ctx.GetPlace()); |
| } | ||
| }; | ||
| out->mutable_data<T>(dev_ctx.GetPlace()); | ||
| paddle::framework::VisitDataType( |
There was a problem hiding this comment.
这里感觉可以用PD_VISIT_ALL_TYPES替换framework::VisitDataType，这样就不用转成proto::VarType了
| #include "paddle/fluid/platform/device/gpu/gpu_primitives.h" | ||
| #include "paddle/phi/core/kernel_registry.h" | ||
| #include "paddle/phi/kernels/funcs/math_function.h" | ||
| #include "paddle/phi/kernels/one_hot_kernel.h" |
| void apply() const { | ||
| auto* p_in_data = in_->data<InT>(); | ||
| auto numel = in_->numel(); | ||
| auto* p_out_data = out_->mutable_data<OutT>(ctx_.GetPlace()); |
|
|
||
| #pragma once | ||
|
|
||
| #include "paddle/phi/common/scalar.h" |
chenwhql
reviewed
Feb 25, 2022
| @@ -477,6 +477,22 @@ phi::InferMetaContext BuildInferMetaContext(InferShapeContext* ctx, | |||
| "Unsupported attribute type is received when call " | |||
| "InferShapeFunctor.")); | |||
| } | |||
| } else if (ctx->HasInput(attr_name)) { | |||
There was a problem hiding this comment.
既然scalar去不掉的话，我们是否有必要单独为depth增加这几处分支
| }; | ||
| out->mutable_data<T>(dev_ctx.GetPlace()); | ||
| paddle::framework::VisitDataType( | ||
| static_cast<paddle::framework::proto::VarType::Type>(dtype), |
There was a problem hiding this comment.
这两个值不相等的吧，如果想用这个，core/utils/data_type.h也有phi的VisitDataType，但这里确实有点乱了，需要整理
| out->Resize(out_dims); | ||
| } | ||
| dev_ctx.template Alloc<T>(out); | ||
| paddle::framework::VisitDataType( |
| }; | ||
|
|
||
| template <typename T, typename Context> | ||
| void OneHotKernel(const Context& dev_ctx, |
There was a problem hiding this comment.
命名也直接叫Raw？其实有raw的，也得同时注册下非row的kernel，我们自己的话最好迁全一点
There was a problem hiding this comment.
OneHot是不是应该直接调用OneHotRaw？然后参考math_kernel中的那些，直接在kernels根目录下加kernel
paddle/phi/ops/compat/one_hot_sig.cc
Outdated
|
|
||
| } // namespace phi | ||
|
|
||
| PD_REGISTER_BASE_KERNEL_NAME(one_hot_v2, one_hot_raw); |
There was a problem hiding this comment.
这里映射到raw不太好，我们有个隐含的原则，有xxx_raw，必有xxx，不能只有raw kernel，不然raw kernel就应该直接注册为非raw的
… move_one_hot_to_phi
… move_one_hot_to_phi
… move_one_hot_to_phi
… move_one_hot_to_phi
… move_one_hot_to_phi
… move_one_hot_to_phi
chenwhql
reviewed
Mar 11, 2022
| void OneHotRawKernel(const Context& dev_ctx, | ||
| const DenseTensor& x, | ||
| int32_t depth, | ||
| int dtype, |
… move_one_hot_to_phi
… move_one_hot_to_phi
… move_one_hot_to_phi
… move_one_hot_to_phi
… into move_one_hot_to_phi
… into move_one_hot_to_phi
chenwhql
reviewed
Mar 14, 2022
| #include "paddle/fluid/platform/device/gpu/gpu_primitives.h" | ||
| #include "paddle/phi/core/kernel_registry.h" | ||
| #include "paddle/phi/kernels/funcs/math_function.h" | ||
| #include "paddle/phi/kernels/one_hot_kernel.h" |
| }; | ||
|
|
||
| template <typename T, typename Context> | ||
| void OneHotKernel(const Context& dev_ctx, |
There was a problem hiding this comment.
OneHot是不是应该直接调用OneHotRaw？然后参考math_kernel中的那些，直接在kernels根目录下加kernel
| } | ||
|
|
||
| template <typename T, typename Context> | ||
| void OneHotKernel(const Context& dev_ctx, |
| @@ -477,6 +477,22 @@ phi::InferMetaContext BuildInferMetaContext(InferShapeContext* ctx, | |||
| "Unsupported attribute type is received when call " | |||
| "InferShapeFunctor.")); | |||
| } | |||
| } else if (ctx->HasInput(attr_name)) { | |||
There was a problem hiding this comment.
既然scalar去不掉的话，我们是否有必要单独为depth增加这几处分支
… move_one_hot_to_phi
… move_one_hot_to_phi
chenwhql
approved these changes
Mar 15, 2022
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
Breaking changes
PR changes
OPs
Describe
move one hot to phi