PaddlePaddle
diff --git a/‎paddle/fluid/distributed/collective/reducer.cc
Copy file name to clipboard
+2-2Lines changed: 2 additions & 2 deletions b/‎paddle/fluid/distributed/collective/reducer.cc
Copy file name to clipboard
+2-2Lines changed: 2 additions & 2 deletions
diff --git a/‎paddle/fluid/distributed/collective/reducer.h
Copy file name to clipboard
+1-1Lines changed: 1 addition & 1 deletion b/‎paddle/fluid/distributed/collective/reducer.h
Copy file name to clipboard
+1-1Lines changed: 1 addition & 1 deletion
diff --git a/‎paddle/fluid/distributed/index_dataset/index_sampler.h
Copy file name to clipboard
+4-5Lines changed: 4 additions & 5 deletions b/‎paddle/fluid/distributed/index_dataset/index_sampler.h
Copy file name to clipboard
+4-5Lines changed: 4 additions & 5 deletions
diff --git a/‎paddle/fluid/framework/CMakeLists.txt
Copy file name to clipboard
-1Lines changed: 0 additions & 1 deletion b/‎paddle/fluid/framework/CMakeLists.txt
Copy file name to clipboard
-1Lines changed: 0 additions & 1 deletion
diff --git a/‎paddle/fluid/imperative/reducer.cc
Copy file name to clipboard
+4-5Lines changed: 4 additions & 5 deletions b/‎paddle/fluid/imperative/reducer.cc
Copy file name to clipboard
+4-5Lines changed: 4 additions & 5 deletions
diff --git a/‎paddle/fluid/operators/CMakeLists.txt
Copy file name to clipboard
+1-1Lines changed: 1 addition & 1 deletion b/‎paddle/fluid/operators/CMakeLists.txt
Copy file name to clipboard
+1-1Lines changed: 1 addition & 1 deletion
diff --git a/‎paddle/fluid/operators/array_to_lod_tensor_op.cc
Copy file name to clipboard
+2-2Lines changed: 2 additions & 2 deletions b/‎paddle/fluid/operators/array_to_lod_tensor_op.cc
Copy file name to clipboard
+2-2Lines changed: 2 additions & 2 deletions
diff --git a/‎paddle/fluid/operators/collective/c_concat_op.cu.cc
Copy file name to clipboard
+2-2Lines changed: 2 additions & 2 deletions b/‎paddle/fluid/operators/collective/c_concat_op.cu.cc
Copy file name to clipboard
+2-2Lines changed: 2 additions & 2 deletions
diff --git a/‎paddle/fluid/operators/detection/collect_fpn_proposals_op.cu
Copy file name to clipboard
+1-1Lines changed: 1 addition & 1 deletion b/‎paddle/fluid/operators/detection/collect_fpn_proposals_op.cu
Copy file name to clipboard
+1-1Lines changed: 1 addition & 1 deletion
diff --git a/‎paddle/fluid/operators/lod_tensor_to_array_op.cc
Copy file name to clipboard
+2-2Lines changed: 2 additions & 2 deletions b/‎paddle/fluid/operators/lod_tensor_to_array_op.cc
Copy file name to clipboard
+2-2Lines changed: 2 additions & 2 deletions
@@ -162,7 +162,7 @@ struct ConcatTensorsForAllReduce {
   void operator()(const DeviceContext &context,
                   const std::vector<phi::DenseTensor> &dense_tensors_,
                   Tensor *p_dense_contents) {
-    operators::math::ConcatFunctor<DeviceContext, T> concat_functor_;
+    phi::funcs::ConcatFunctor<DeviceContext, T> concat_functor_;
     concat_functor_(
         context,
         dense_tensors_,
@@ -191,7 +191,7 @@ struct SplitTensorsForAllReduce {
       shape_refer.emplace_back(&tensor);
     }
 
-    operators::math::SplitFunctor<DeviceContext, T> split_functor_;
+    phi::funcs::SplitFunctor<DeviceContext, T> split_functor_;
     split_functor_(context, *in, shape_refer, 0, &outs);
   }
 };
 
@@ -22,12 +22,12 @@
 #include "paddle/fluid/eager/api/utils/hook_utils.h"
 #include "paddle/fluid/eager/autograd_meta.h"
 #include "paddle/fluid/eager/utils.h"
-#include "paddle/fluid/operators/math/concat_and_split.h"
 #include "paddle/fluid/platform/device/gpu/gpu_info.h"
 #include "paddle/phi/api/include/api.h"
 #include "paddle/phi/api/include/fused_api.h"
 #include "paddle/phi/api/include/tensor.h"
 #include "paddle/phi/common/data_type.h"
+#include "paddle/phi/kernels/funcs/concat_and_split_functor.h"
 #include "paddle/phi/kernels/funcs/math_function.h"
 #include "paddle/utils/string/string_helper.h"
 
 
@@ -18,8 +18,8 @@
 #include "paddle/fluid/distributed/index_dataset/index_wrapper.h"
 #include "paddle/fluid/framework/data_feed.h"
 #include "paddle/fluid/framework/program_desc.h"
-#include "paddle/fluid/operators/math/sampler.h"
 #include "paddle/fluid/platform/enforce.h"
+#include "paddle/phi/kernels/funcs/math/sampler.h"
 
 namespace paddle {
 namespace distributed {
@@ -107,9 +107,8 @@ class LayerWiseSampler : public IndexSampler {
     while (layer_index >= start_sample_layer_) {
       auto layer_codes = tree_->GetLayerCodes(layer_index);
       layer_ids_.push_back(tree_->GetNodes(layer_codes));
-      auto sampler_temp =
-          std::make_shared<paddle::operators::math::UniformSampler>(
-              layer_ids_[idx].size() - 1, seed_);
+      auto sampler_temp = std::make_shared<phi::math::UniformSampler>(
+          layer_ids_[idx].size() - 1, seed_);
       sampler_vec_.push_back(sampler_temp);
       layer_index--;
       idx++;
@@ -131,7 +130,7 @@ class LayerWiseSampler : public IndexSampler {
   std::shared_ptr<TreeIndex> tree_{nullptr};
   int seed_{0};
   int start_sample_layer_{1};
-  std::vector<std::shared_ptr<paddle::operators::math::Sampler>> sampler_vec_;
+  std::vector<std::shared_ptr<phi::math::Sampler>> sampler_vec_;
   std::vector<std::vector<IndexNode>> layer_ids_;
 };
 
 
@@ -671,7 +671,6 @@ if(WITH_DISTRIBUTE)
            glog
            index_sampler
            index_wrapper
-           sampler
            index_dataset_proto
            lod_rank_table
            framework_io
 
@@ -19,7 +19,7 @@
 #include "paddle/fluid/framework/tensor_util.h"
 #include "paddle/fluid/imperative/layer.h"
 #include "paddle/fluid/imperative/parallel_context.h"
-#include "paddle/fluid/operators/math/concat_and_split.h"
+#include "paddle/phi/kernels/funcs/concat_and_split_functor.h"
 #include "paddle/phi/kernels/funcs/strided_memcpy.h"
 #ifdef PADDLE_WITH_XPU
 #include "paddle/fluid/platform/device/xpu/enforce_xpu.h"
@@ -74,7 +74,7 @@ static void ConcatTensorsForAllReduce(
     const DeviceContext &context,
     const std::vector<phi::DenseTensor> &dense_tensors_,
     framework::Variable *p_dense_contents) {
-  operators::math::ConcatFunctor<DeviceContext, T> concat_functor_;
+  phi::funcs::ConcatFunctor<DeviceContext, T> concat_functor_;
   concat_functor_(context,
                   dense_tensors_,
                   0,
@@ -102,7 +102,7 @@ static void SplitTensorsForAllReduce(
     phi::funcs::StridedMemcpyWithAxis0<T, DeviceContext>(
         context, *in, shape_refer, &outs);
   } else {
-    operators::math::SplitFunctor<DeviceContext, T> split_functor_;
+    phi::funcs::SplitFunctor<DeviceContext, T> split_functor_;
     split_functor_(context, *in, shape_refer, 0, &outs);
   }
 }
@@ -179,8 +179,7 @@ void SplitTensorsForAllReduce<platform::XPUDeviceContext, float>(
     outs.emplace_back(&tensor);
     shape_refer.emplace_back(&tensor);
   }
-  operators::math::SplitFunctor<platform::XPUDeviceContext, float>
-      split_functor_;
+  phi::funcs::SplitFunctor<platform::XPUDeviceContext, float> split_functor_;
   split_functor_(context, *in, shape_refer, 0, &outs);
 }
 
 
@@ -134,7 +134,7 @@ set(COMMON_OP_DEPS ${COMMON_OP_DEPS} phi common)
 set(COMMON_OP_DEPS ${COMMON_OP_DEPS} selected_rows_utils
 lod_tensor unpooling lod_rank_table context_project executor static_prim_api)
 set(COMMON_OP_DEPS ${COMMON_OP_DEPS} dynload_warpctc static_prim_api static_utils static_global_utils prim_utils)
-set(COMMON_OP_DEPS ${COMMON_OP_DEPS} cos_sim_functor    concat_and_split sampler sample_prob tree2col)
+set(COMMON_OP_DEPS ${COMMON_OP_DEPS} cos_sim_functor    concat_and_split tree2col)
 set(COMMON_OP_DEPS ${COMMON_OP_DEPS} beam_search)
 set(COMMON_OP_DEPS ${COMMON_OP_DEPS} box_wrapper ps_gpu_wrapper)
 set(COMMON_OP_DEPS ${COMMON_OP_DEPS} common_infer_shape_functions)
 
@@ -11,7 +11,7 @@ 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/math/concat_and_split.h>
+#include "paddle/phi/kernels/funcs/concat_and_split_functor.h"
 
 #include "paddle/fluid/framework/op_registry.h"
 #include "paddle/fluid/platform/device_context.h"
@@ -77,7 +77,7 @@ struct ArrayToLoDFunctor {
 template <typename DeviceContext>
 template <typename T>
 void ArrayToLoDFunctorImpl<DeviceContext>::apply() {
-  math::ConcatFunctor<DeviceContext, T> func;
+  phi::funcs::ConcatFunctor<DeviceContext, T> func;
   func(*dev_ctx_, prev_functor_->in, 0, prev_functor_->out);
 }
 
 
@@ -17,8 +17,8 @@ limitations under the License. */
 #include <vector>
 #include "paddle/phi/core/distributed/comm_context_manager.h"
 
-#include "paddle/fluid/operators/math/concat_and_split.h"
 #include "paddle/phi/api/include/tensor.h"
+#include "paddle/phi/kernels/funcs/concat_and_split_functor.h"
 
 #if defined(PADDLE_WITH_NCCL) || defined(PADDLE_WITH_RCCL)
 #include "paddle/common/flags.h"
@@ -151,7 +151,7 @@ class CConcatOpCUDAKernel : public framework::OpKernel<T> {
       offset += rows_per_tensor;
     }
 
-    math::ConcatFunctor<phi::GPUContext, T> functor;
+    phi::funcs::ConcatFunctor<phi::GPUContext, T> functor;
     out->mutable_data<T>(out_dims, place);
     auto& dev_ctx2 = ctx.template device_context<phi::GPUContext>();
     functor(dev_ctx2, inputs, axis, out);
 
@@ -23,10 +23,10 @@ namespace cub = hipcub;
 #include "paddle/fluid/memory/memcpy.h"
 #include "paddle/fluid/operators/detection/bbox_util.h"
 #include "paddle/fluid/operators/detection/collect_fpn_proposals_op.h"
-#include "paddle/fluid/operators/math/concat_and_split.h"
 #include "paddle/fluid/platform/for_range.h"
 #include "paddle/phi/backends/gpu/gpu_primitives.h"
 #include "paddle/phi/core/mixed_vector.h"
+#include "paddle/phi/kernels/funcs/concat_and_split_functor.h"
 #include "paddle/phi/kernels/funcs/gather.cu.h"
 #include "paddle/phi/kernels/funcs/strided_memcpy.h"
 
 
@@ -12,9 +12,9 @@ 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/framework/op_registry.h"
-#include "paddle/fluid/operators/math/concat_and_split.h"
 #include "paddle/fluid/platform/device_context.h"
 #include "paddle/phi/core/lod_utils.h"
+#include "paddle/phi/kernels/funcs/concat_and_split_functor.h"
 
 namespace paddle {
 namespace framework {
@@ -88,7 +88,7 @@ struct LoDTensorToArrayFunctor {
 template <typename DeviceContext>
 template <typename T>
 void LoDTensorToArrayFunctorImpl<DeviceContext>::apply() {
-  math::SplitFunctor<DeviceContext, T> func;
+  phi::funcs::SplitFunctor<DeviceContext, T> func;
   func(*dev_ctx_,
        prev_functor_->input_,
        prev_functor_->ref_inputs_,