[phi] Move sequence_pool to phi - Step 3 ：sequence_pool_grad_op (#52680)

* [phi] move sequence_pool kernel to phi

* mv kernels impl

* fix parameter error

* clean include

* fix compat filename

* [phi] move fluid sequence_pool_grad to phi

* [phi][compat] sig rm GradVarName

* [phi] fix sequence_pool out type

* [phi] rm impl, add const string

* [phi] fix const str

* fix sequence_pooling cmake

* [phi] mv sequence_pooling_test

* [phi] fix grad sig

* [phi] fix sequence_pool is_test error

* [phi] fix sequence_pooling gpu include

* [phi] mv to impl

* [phi] fix SequencePoolFunctor cu include

* [phi] modify out max_index int32_t

* [phi] add pooltype mapping determine

* [phi] fix sequence_pool_sig

* [phi] fix sequence_pool_sig sum

* [phi] try ci

* [phi] fix max_index optional

paddle/fluid/operators/math/CMakeLists.txt

@@ -14,7 +14,6 @@ math_library(sample_prob)
 math_library(sampler DEPS generator)
 
 # math_library(math_function DEPS blas dense_tensor tensor)
-
 if(WITH_XPU)
   math_library(beam_search DEPS math_function beam_search_xpu)
 else()

paddle/fluid/operators/sequence_ops/sequence_pool_op.cc

@@ -12,10 +12,7 @@ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 See the License for the specific language governing permissions and
 limitations under the License. */
 
-#include "paddle/fluid/operators/sequence_ops/sequence_pool_op.h"
-
-#include <memory>
-#include <string>
+#include "paddle/fluid/framework/op_registry.h"
 
 namespace paddle {
 namespace operators {
@@ -196,10 +193,3 @@ REGISTER_OPERATOR(sequence_pool,
 REGISTER_OPERATOR(sequence_pool_grad,
                   ops::SequencePoolGradOp,
                   ops::SequencePoolGradOpNoNeedBufferVarsInferer);
-
-PD_REGISTER_STRUCT_KERNEL(sequence_pool_grad,
-                          CPU,
-                          ALL_LAYOUT,
-                          ops::SequencePoolGradKernel,
-                          float,
-                          double) {}

paddle/fluid/operators/sequence_ops/unity_build_rule.cmake

@@ -30,7 +30,6 @@ register_unity_group(
   sequence_expand_op.cu
   sequence_mask_op.cu
   sequence_pad_op.cu
-  sequence_pool_op.cu
   sequence_expand_as_op.cu
   sequence_reshape_op.cu
   sequence_reverse_op.cu

paddle/phi/kernels/cpu/sequence_pool_grad_kernel.cc

+/* Copyright (c) 2023 PaddlePaddle Authors. All Rights Reserved.
+
+Licensed under the Apache License, Version 2.0 (the "License");
+you may not use this file except in compliance with the License.
+You may obtain a copy of the License at
+
+    http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software
+distributed under the License is distributed on an "AS IS" BASIS,
+WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+See the License for the specific language governing permissions and
+limitations under the License. */
+
+#include "paddle/phi/kernels/sequence_pool_grad_kernel.h"
+
+#include "paddle/phi/backends/cpu/cpu_context.h"
+#include "paddle/phi/core/kernel_registry.h"
+#include "paddle/phi/kernels/impl/sequence_pool_grad_kernel_impl.h"
+
+PD_REGISTER_KERNEL(sequence_pool_grad,
+                   CPU,
+                   ALL_LAYOUT,
+                   phi::SequencePoolGradKernel,
+                   float,
+                   double) {}

paddle/fluid/operators/sequence_ops/sequence_pool_op.cu → paddle/phi/kernels/gpu/sequence_pool_grad_kernel.cu

-/* Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserved.
+/* Copyright (c) 2023 PaddlePaddle Authors. All Rights Reserved.
 
 Licensed under the Apache License, Version 2.0 (the "License");
 you may not use this file except in compliance with the License.
@@ -11,8 +11,12 @@ distributed under the License is distributed on an "AS IS" BASIS,
 WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 See the License for the specific language governing permissions and
 limitations under the License. */
-#include "paddle/fluid/operators/sequence_ops/sequence_pool_op.h"
 
-namespace ops = paddle::operators;
-PD_REGISTER_STRUCT_KERNEL(
-    sequence_pool_grad, GPU, ALL_LAYOUT, ops::SequencePoolGradKernel, float) {}
+#include "paddle/phi/kernels/sequence_pool_grad_kernel.h"
+#include "paddle/phi/kernels/impl/sequence_pool_grad_kernel_impl.h"
+
+#include "paddle/phi/backends/gpu/gpu_context.h"
+#include "paddle/phi/core/kernel_registry.h"
+
+PD_REGISTER_KERNEL(
+    sequence_pool_grad, GPU, ALL_LAYOUT, phi::SequencePoolGradKernel, float) {}

paddle/fluid/operators/sequence_ops/sequence_pool_op.h → paddle/phi/kernels/impl/sequence_pool_grad_kernel_impl.h

-/* Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserved.
+/* Copyright (c) 2023 PaddlePaddle Authors. All Rights Reserved.
 
 Licensed under the Apache License, Version 2.0 (the "License");
 you may not use this file except in compliance with the License.
@@ -13,37 +13,27 @@ See the License for the specific language governing permissions and
 limitations under the License. */
 
 #pragma once
-#include <string>
 
-#include "paddle/fluid/framework/eigen.h"
-#include "paddle/fluid/framework/op_registry.h"
-#include "paddle/phi/kernels/funcs/math_function.h"
 #include "paddle/phi/kernels/funcs/sequence_pooling.h"
 
-namespace paddle {
-namespace operators {
-
-template <typename T, typename DeviceContext>
-class SequencePoolGradKernel : public framework::OpKernel<T> {
- public:
-  void Compute(const framework::ExecutionContext& context) const override {
-    auto* out_g =
-        context.Input<phi::DenseTensor>(framework::GradVarName("Out"));
-    auto* in_g = context.Output<phi::DenseTensor>(framework::GradVarName("X"));
-    std::string pooltype = context.Attr<std::string>("pooltype");
-    const phi::DenseTensor* index = nullptr;
-    if (pooltype == "MAX") {
-      index = context.Input<phi::DenseTensor>("MaxIndex");
-    }
-    in_g->mutable_data<T>(context.GetPlace());
-    phi::funcs::SequencePoolGradFunctor<DeviceContext, T> pool;
-    pool(context.template device_context<DeviceContext>(),
-         pooltype,
-         *out_g,
-         in_g,
-         index);
+namespace phi {
+
+template <typename T, typename Context>
+void SequencePoolGradKernel(const Context& dev_ctx,
+                            const DenseTensor& x,
+                            const paddle::optional<DenseTensor>& max_index,
+                            const DenseTensor& out_grad,
+                            bool is_test,
+                            const std::string& pooltype,
+                            float pad_value,
+                            DenseTensor* x_grad) {
+  const phi::DenseTensor* index = nullptr;
+  if (pooltype == "MAX") {
+    index = max_index.get_ptr();
   }
-};
+  dev_ctx.template Alloc<T>(x_grad);
+  phi::funcs::SequencePoolGradFunctor<Context, T> pool;
+  pool(dev_ctx, pooltype, out_grad, x_grad, index);
+}
 
-}  // namespace operators
-}  // namespace paddle
+}  // namespace phi

paddle/phi/kernels/impl/sequence_pool_kernel_impl.h

@@ -68,7 +68,7 @@ void SequencePoolKernel(const Context& ctx,
       (is_test == false || (ctx.GetPlace() == phi::CPUPlace()) == false)) {
     index = max_index;
     index->Resize({dims});
-    ctx.template Alloc<int>(index);
+    ctx.template Alloc<int32_t>(index);
   }
   phi::funcs::SequencePoolFunctor<Context, T> pool;
   pool(ctx, pooltype, pad_value_, x, out, is_test, index);

paddle/phi/kernels/sequence_pool_grad_kernel.h

+/* Copyright (c) 2023 PaddlePaddle Authors. All Rights Reserved.
+
+Licensed under the Apache License, Version 2.0 (the "License");
+you may not use this file except in compliance with the License.
+You may obtain a copy of the License at
+
+    http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software
+distributed under the License is distributed on an "AS IS" BASIS,
+WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+See the License for the specific language governing permissions and
+limitations under the License. */
+
+#pragma once
+
+#include "paddle/phi/core/dense_tensor.h"
+
+namespace phi {
+template <typename T, typename Context>
+void SequencePoolGradKernel(const Context& dev_ctx,
+                            const DenseTensor& x,
+                            const paddle::optional<DenseTensor>& max_index,
+                            const DenseTensor& out_grad,
+                            bool is_test,
+                            const std::string& pooltype,
+                            float pad_value,
+                            DenseTensor* x_grad);
+
+}  // namespace phi

paddle/phi/ops/compat/sequence_pool_sig.cc

@@ -21,6 +21,16 @@ KernelSignature SequencePoolOpArgumentMapping(
                          {"Out", "MaxIndex"});
 }
 
+KernelSignature SequencePoolGradOpArgumentMapping(
+    const ArgumentMappingContext& ctx) {
+  return KernelSignature("sequence_pool_grad",
+                         {"X", "MaxIndex", "Out@GRAD"},
+                         {"is_test", "pooltype", "pad_value"},
+                         {"X@GRAD"});
+}
+
 }  // namespace phi
 
 PD_REGISTER_ARG_MAPPING_FN(sequence_pool, phi::SequencePoolOpArgumentMapping);
+PD_REGISTER_ARG_MAPPING_FN(sequence_pool_grad,
+                           phi::SequencePoolGradOpArgumentMapping);