Merge pull request #13689 from sneaxiy/sparse_rmsprop

Fix sparse rmsprop

paddle/fluid/operators/adam_op.h

@@ -18,6 +18,7 @@ limitations under the License. */
 #include <vector>
 #include "paddle/fluid/framework/op_registry.h"
 #include "paddle/fluid/operators/detail/safe_ref.h"
+#include "paddle/fluid/operators/math/algorithm.h"
 #include "paddle/fluid/operators/math/selected_rows_functor.h"
 #include "paddle/fluid/platform/for_range.h"
 
@@ -199,23 +200,9 @@ struct SparseAdamFunctor {
         row_numel_(row_numel),
         row_count_(row_count) {}
 
-  inline HOSTDEVICE int64_t BinarySearchInRows(int64_t row) const {
-    int64_t beg = 0, end = row_count_ - 1;
-    while (beg <= end) {
-      auto mid = ((beg + end) >> 1);
-      if (rows_[mid] == row)
-        return mid;
-      else if (rows_[mid] < row)
-        beg = mid + 1;
-      else
-        end = mid - 1;
-    }
-    return -1;
-  }
-
   inline HOSTDEVICE void operator()(size_t i) const {
-    int64_t row = i / row_numel_;
-    auto row_idx = BinarySearchInRows(row);
+    auto row_idx =
+        math::BinarySearch<int64_t>(rows_, row_count_, i / row_numel_);
     T g = row_idx >= 0 ? grad_[row_idx * row_numel_ + i % row_numel_] : 0;
 
     // The following code is the same as dense

paddle/fluid/operators/math/algorithm.h

+// Copyright (c) 2018 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 <algorithm>
+#include <cstdint>  // for int64_t
+#include <numeric>
+
+#include "paddle/fluid/platform/hostdevice.h"
+
+namespace paddle {
+namespace operators {
+namespace math {
+
+template <typename T>
+HOSTDEVICE inline int64_t BinarySearch(const T *x, int64_t num, const T &val) {
+  int64_t beg = 0, end = num - 1;
+  while (beg <= end) {
+    auto mid = ((beg + end) >> 1);
+    if (x[mid] == val)
+      return mid;
+    else if (x[mid] < val)
+      beg = mid + 1;
+    else
+      end = mid - 1;
+  }
+  return -1;
+}
+
+}  // namespace math
+}  // namespace operators
+}  // namespace paddle

paddle/fluid/operators/math/selected_rows_functor.cc

@@ -12,9 +12,11 @@ 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 <map>
 #include <set>
 #include <vector>
 
+#include "paddle/fluid/operators/math/blas.h"
 #include "paddle/fluid/operators/math/selected_rows_functor.h"
 
 namespace paddle {
@@ -245,40 +247,42 @@ struct MergeAdd<platform::CPUDeviceContext, T> {
                   const framework::SelectedRows& input,
                   framework::SelectedRows* output) {
     framework::SelectedRows& out = *output;
-    auto input_rows = input.rows();
-    std::vector<int64_t> merge_rows;
-    merge_rows.reserve(input_rows.size());
-    std::unordered_map<int64_t, size_t> rows_pos_map;
-    rows_pos_map.reserve(input_rows.size());
-    size_t idx = 0u;
-    for (std::vector<int64_t>::iterator iter = input_rows.begin();
-         iter != input_rows.end(); ++iter) {
-      if (rows_pos_map.find(*iter) == rows_pos_map.end()) {
-        rows_pos_map[*iter] = idx++;
-        merge_rows.emplace_back(*iter);
-      }
+    std::vector<int64_t> input_rows(input.rows());
+
+    std::map<int64_t, std::vector<int64_t>> merge_row_map;
+    for (size_t i = 0; i < input_rows.size(); ++i) {
+      merge_row_map[input_rows[i]].push_back(i);
     }
 
-    auto input_width = input.value().dims()[1];
-    out.set_rows(merge_rows);
+    std::vector<int64_t> merge_rows(merge_row_map.size());
+    size_t idx = 0;
+    int64_t input_width = input.value().dims()[1];
     out.set_height(input.height());
-    out.mutable_value()->mutable_data<T>(
+
+    T* out_data = out.mutable_value()->mutable_data<T>(
         framework::make_ddim(
             {static_cast<int64_t>(merge_rows.size()), input_width}),
         context.GetPlace());
-
-    math::SetConstant<platform::CPUDeviceContext, T> constant_functor;
-    constant_functor(context, out.mutable_value(), 0.0);
-
-    auto* out_data = out.mutable_value()->data<T>();
-    auto* input_data = input.value().data<T>();
-
-    for (size_t i = 0; i < input_rows.size(); i++) {
-      size_t out_i = rows_pos_map[input_rows[i]];
-      for (int64_t j = 0; j < input_width; j++) {
-        out_data[out_i * input_width + j] += input_data[i * input_width + j];
+    const T* in_data = input.value().data<T>();
+
+    for (auto& row_pair : merge_row_map) {
+      auto* out_ptr = out_data + idx * input_width;
+      auto& rows = row_pair.second;
+      merge_rows[idx] = row_pair.first;
+      ++idx;
+      // rows.size() is always larger than 0
+      std::memcpy(out_ptr, in_data + rows[0] * input_width,
+                  sizeof(T) * input_width);
+
+      for (size_t i = 1; i < rows.size(); ++i) {
+        auto* in_ptr = in_data + rows[i] * input_width;
+        for (int64_t j = 0; j < input_width; ++j) {
+          out_ptr[j] += in_ptr[j];
+        }
       }
     }
+
+    out.set_rows(merge_rows);
   }
 };
 

paddle/fluid/operators/math/selected_rows_functor.h

@@ -13,6 +13,7 @@ See the License for the specific language governing permissions and
 limitations under the License. */
 #pragma once
 
+#include <map>
 #include <vector>
 
 #include "paddle/fluid/framework/eigen.h"
@@ -97,41 +98,39 @@ struct MergeAdd<platform::CPUDeviceContext, float> {
                   const framework::SelectedRows& input,
                   framework::SelectedRows* output) {
     framework::SelectedRows& out = *output;
-    auto input_rows = input.rows();
-    std::vector<int64_t> merge_rows;
-    merge_rows.reserve(input_rows.size());
-    std::unordered_map<int64_t, size_t> rows_pos_map;
-    rows_pos_map.reserve(input_rows.size());
-    size_t idx = 0u;
-    for (std::vector<int64_t>::iterator iter = input_rows.begin();
-         iter != input_rows.end(); ++iter) {
-      if (rows_pos_map.find(*iter) == rows_pos_map.end()) {
-        rows_pos_map[*iter] = idx++;
-        merge_rows.emplace_back(*iter);
-      }
+    std::vector<int64_t> input_rows(input.rows());
+
+    std::map<int64_t, std::vector<int64_t>> merge_row_map;
+    for (size_t i = 0; i < input_rows.size(); ++i) {
+      merge_row_map[input_rows[i]].push_back(i);
     }
 
-    auto input_width = input.value().dims()[1];
-    out.set_rows(merge_rows);
+    std::vector<int64_t> merge_rows(merge_row_map.size());
+    size_t idx = 0;
+    int64_t input_width = input.value().dims()[1];
     out.set_height(input.height());
-    out.mutable_value()->mutable_data<float>(
+
+    auto* out_data = out.mutable_value()->mutable_data<float>(
         framework::make_ddim(
             {static_cast<int64_t>(merge_rows.size()), input_width}),
         context.GetPlace());
-
-    math::SetConstant<platform::CPUDeviceContext, float> constant_functor;
-    constant_functor(context, out.mutable_value(), 0.0);
-
-    auto* out_data = out.mutable_value()->data<float>();
-    auto* input_data = input.value().data<float>();
+    auto* in_data = input.value().data<float>();
 
     auto blas = GetBlas<platform::CPUDeviceContext, float>(context);
-    for (size_t i = 0; i < input_rows.size(); i++) {
-      size_t out_i = rows_pos_map[input_rows[i]];
-      float* y = out_data + out_i * input_width;
-      const float* x = input_data + i * input_width;
-      blas.AXPY(input_width, 1., x, y);
+    for (auto& row_pair : merge_row_map) {
+      auto* out_ptr = out_data + idx * input_width;
+      auto& rows = row_pair.second;
+      merge_rows[idx] = row_pair.first;
+      ++idx;
+      // rows.size() is always larger than 0
+      blas.VCOPY(input_width, in_data + rows[0] * input_width, out_ptr);
+
+      for (size_t i = 1; i < rows.size(); ++i) {
+        blas.AXPY(input_width, 1., in_data + rows[i] * input_width, out_ptr);
+      }
     }
+
+    out.set_rows(merge_rows);
   }
 };
 
@@ -148,41 +147,39 @@ struct MergeAdd<platform::CPUDeviceContext, double> {
                   const framework::SelectedRows& input,
                   framework::SelectedRows* output) {
     framework::SelectedRows& out = *output;
-    auto input_rows = input.rows();
-    std::vector<int64_t> merge_rows;
-    merge_rows.reserve(input_rows.size());
-    std::unordered_map<int64_t, size_t> rows_pos_map;
-    rows_pos_map.reserve(input_rows.size());
-    size_t idx = 0u;
-    for (std::vector<int64_t>::iterator iter = input_rows.begin();
-         iter != input_rows.end(); ++iter) {
-      if (rows_pos_map.find(*iter) == rows_pos_map.end()) {
-        rows_pos_map[*iter] = idx++;
-        merge_rows.emplace_back(*iter);
-      }
+    std::vector<int64_t> input_rows(input.rows());
+
+    std::map<int64_t, std::vector<int64_t>> merge_row_map;
+    for (size_t i = 0; i < input_rows.size(); ++i) {
+      merge_row_map[input_rows[i]].push_back(i);
     }
 
-    auto input_width = input.value().dims()[1];
-    out.set_rows(merge_rows);
+    std::vector<int64_t> merge_rows(merge_row_map.size());
+    size_t idx = 0;
+    int64_t input_width = input.value().dims()[1];
     out.set_height(input.height());
-    out.mutable_value()->mutable_data<double>(
+
+    auto* out_data = out.mutable_value()->mutable_data<double>(
         framework::make_ddim(
             {static_cast<int64_t>(merge_rows.size()), input_width}),
         context.GetPlace());
-
-    math::SetConstant<platform::CPUDeviceContext, double> constant_functor;
-    constant_functor(context, out.mutable_value(), 0.0);
-
-    auto* out_data = out.mutable_value()->data<double>();
-    auto* input_data = input.value().data<double>();
+    auto* in_data = input.value().data<double>();
 
     auto blas = GetBlas<platform::CPUDeviceContext, double>(context);
-    for (size_t i = 0; i < input_rows.size(); i++) {
-      size_t out_i = rows_pos_map[input_rows[i]];
-      double* y = out_data + out_i * input_width;
-      const double* x = input_data + i * input_width;
-      blas.AXPY(input_width, 1., x, y);
+    for (auto& row_pair : merge_row_map) {
+      auto* out_ptr = out_data + idx * input_width;
+      auto& rows = row_pair.second;
+      merge_rows[idx] = row_pair.first;
+      ++idx;
+      // rows.size() is always larger than 0
+      blas.VCOPY(input_width, in_data + rows[0] * input_width, out_ptr);
+
+      for (size_t i = 1; i < rows.size(); ++i) {
+        blas.AXPY(input_width, 1., in_data + rows[i] * input_width, out_ptr);
+      }
     }
+
+    out.set_rows(merge_rows);
   }
 };
 

paddle/fluid/operators/rmsprop_op.h

@@ -13,72 +13,254 @@ See the License for the specific language governing permissions and
 limitations under the License. */
 
 #pragma once
+#include <math.h>
 #include "paddle/fluid/framework/eigen.h"
 #include "paddle/fluid/framework/op_registry.h"
+#include "paddle/fluid/operators/math/algorithm.h"
+#include "paddle/fluid/operators/math/selected_rows_functor.h"
+#include "paddle/fluid/platform/for_range.h"
 
 namespace paddle {
 namespace operators {
 
-using Tensor = framework::Tensor;
 template <typename T, int MajorType = Eigen::RowMajor,
           typename IndexType = Eigen::DenseIndex>
 using EigenVector = framework::EigenVector<T, MajorType, IndexType>;
 
+template <typename T>
+struct DenseRmspropGradFunctor {
+  inline explicit DenseRmspropGradFunctor(const T *grad) : grad_(grad) {}
+
+  HOSTDEVICE inline T operator()(int64_t idx) const { return grad_[idx]; }
+
+  const T *grad_;
+};
+
+template <typename T>
+struct SparseRmspropGradFunctor {
+  inline SparseRmspropGradFunctor(const T *grad, const int64_t *rows,
+                                  int64_t row_numel, int64_t row_count)
+      : grad_(grad),
+        rows_(rows),
+        row_numel_(row_numel),
+        row_count_(row_count) {}
+
+  HOSTDEVICE inline T operator()(int64_t idx) const {
+    auto row_idx = math::BinarySearch(rows_, row_count_, idx / row_numel_);
+    return row_idx >= 0 ? grad_[row_idx * row_numel_ + idx % row_numel_] : 0;
+  }
+
+  const T *grad_;
+  const int64_t *rows_;
+  int64_t row_numel_;
+  int64_t row_count_;
+};
+
+template <typename T, typename GradFunctor>
+struct UncenteredRmspropFunctor {
+  UncenteredRmspropFunctor(T *param, T *ms, T *mom, const T *lr, T rho,
+                           T epsilon, T momentum,
+                           const GradFunctor &grad_functor)
+      : param_(param),
+        ms_(ms),
+        mom_(mom),
+        lr_(lr),
+        rho_(rho),
+        epsilon_(epsilon),
+        momentum_(momentum),
+        grad_functor_(grad_functor) {}
+
+  HOSTDEVICE inline void operator()(int64_t idx) const {
+    T g = grad_functor_(idx);
+    T ms_out = rho_ * ms_[idx] + (1 - rho_) * g * g;
+    T mom_out = momentum_ * mom_[idx] + lr_[0] * g / sqrt(ms_out + epsilon_);
+    param_[idx] -= mom_out;
+    ms_[idx] = ms_out;
+    mom_[idx] = mom_out;
+  }
+
+  T *param_;
+  T *ms_;
+  T *mom_;
+  const T *lr_;
+  T rho_;
+  T epsilon_;
+  T momentum_;
+  GradFunctor grad_functor_;
+};
+
+template <typename T, typename GradFunctor>
+struct CenteredRmspropFunctor {
+  CenteredRmspropFunctor(T *param, T *ms, T *mom, T *mean_grad, const T *lr,
+                         T rho, T epsilon, T momentum,
+                         const GradFunctor &grad_functor)
+      : param_(param),
+        ms_(ms),
+        mom_(mom),
+        mean_grad_(mean_grad),
+        lr_(lr),
+        rho_(rho),
+        epsilon_(epsilon),
+        momentum_(momentum),
+        grad_functor_(grad_functor) {}
+
+  HOSTDEVICE inline void operator()(int64_t idx) const {
+    T g = grad_functor_(idx);
+    T ms_out = rho_ * ms_[idx] + (1 - rho_) * g * g;
+    T mg_out = rho_ * mean_grad_[idx] + (1 - rho_) * g;
+    T mom_out = momentum_ * mom_[idx] +
+                lr_[0] * g / sqrt(ms_out - mg_out * mg_out + epsilon_);
+    param_[idx] -= mom_out;
+    ms_[idx] = ms_out;
+    mom_[idx] = mom_out;
+    mean_grad_[idx] = mg_out;
+  }
+
+  T *param_;
+  T *ms_;
+  T *mom_;
+  T *mean_grad_;
+  const T *lr_;
+  T rho_;
+  T epsilon_;
+  T momentum_;
+  GradFunctor grad_functor_;
+};
+
 template <typename DeviceContext, typename T>
 class RmspropOpKernel : public framework::OpKernel<T> {
  public:
-  void Compute(const framework::ExecutionContext& ctx) const override {
-    const auto* param_var = ctx.InputVar("Param");
-    PADDLE_ENFORCE(param_var->IsType<framework::LoDTensor>(),
-                   "The Var(%s)'s type should be LoDTensor, "
-                   "but the received is %s",
-                   ctx.Inputs("Param").front(), param_var->Type().name());
-
-    auto* param_out = ctx.Output<Tensor>("ParamOut");
-    auto* moment_out = ctx.Output<Tensor>("MomentOut");
-    auto* mean_square_out = ctx.Output<Tensor>("MeanSquareOut");
-
-    auto grad = ctx.Input<Tensor>("Grad");
-
-    param_out->mutable_data<T>(ctx.GetPlace());
-    moment_out->mutable_data<T>(ctx.GetPlace());
-    mean_square_out->mutable_data<T>(ctx.GetPlace());
-
-    float epsilon = ctx.Attr<float>("epsilon");
-    float rho = ctx.Attr<float>("decay");
-    float momentum = ctx.Attr<float>("momentum");
+  void Compute(const framework::ExecutionContext &ctx) const override {
+    using LoDTensor = framework::LoDTensor;
+    auto *grad_var = ctx.InputVar("Grad");
+    auto *param_out = ctx.Output<LoDTensor>("ParamOut");
+    auto *moment_out = ctx.Output<LoDTensor>("MomentOut");
+    auto *mean_square_out = ctx.Output<LoDTensor>("MeanSquareOut");
+
+    auto epsilon = static_cast<T>(ctx.Attr<float>("epsilon"));
+    auto rho = static_cast<T>(ctx.Attr<float>("decay"));
+    auto momentum = static_cast<T>(ctx.Attr<float>("momentum"));
     bool centered = ctx.Attr<bool>("centered");
 
-    auto p = EigenVector<T>::Flatten(*ctx.Input<Tensor>("Param"));
-    auto ms = EigenVector<T>::Flatten(*ctx.Input<Tensor>("MeanSquare"));
-    auto lr = EigenVector<T>::Flatten(*ctx.Input<Tensor>("LearningRate"));
-    auto g = EigenVector<T>::Flatten(*grad);
-    auto mom = EigenVector<T>::Flatten(*ctx.Input<Tensor>("Moment"));
+    auto &p_tensor = *ctx.Input<LoDTensor>("Param");
+    auto &ms_tensor = *ctx.Input<LoDTensor>("MeanSquare");
+    auto &lr_tensor = *ctx.Input<LoDTensor>("LearningRate");
+    auto &mom_tensor = *ctx.Input<LoDTensor>("Moment");
+
+    PADDLE_ENFORCE_EQ(&p_tensor, param_out,
+                      "Param and ParamOut must be the same Tensor");
+    PADDLE_ENFORCE_EQ(&mom_tensor, moment_out,
+                      "Moment and MomentOut must be the same Tensor");
+    PADDLE_ENFORCE_EQ(&ms_tensor, mean_square_out,
+                      "MeanSquare and MeanSquareOut must be the same Tensor");
+
+    auto &dev_ctx = ctx.template device_context<DeviceContext>();
+    size_t limit = static_cast<size_t>(ms_tensor.numel());
+
+    if (grad_var->IsType<LoDTensor>()) {
+      auto &grad_tensor = grad_var->Get<LoDTensor>();
+
+      if (std::is_same<DeviceContext, platform::CPUDeviceContext>::value) {
+        auto &place =
+            *ctx.template device_context<DeviceContext>().eigen_device();
+        auto lr_value = lr_tensor.data<T>()[0];
+
+        auto p = EigenVector<T>::Flatten(p_tensor);
+        auto ms = EigenVector<T>::Flatten(ms_tensor);
+        auto g = EigenVector<T>::Flatten(grad_tensor);
+        auto mom = EigenVector<T>::Flatten(mom_tensor);
 
         auto p_out = EigenVector<T>::Flatten(*param_out);
         auto mom_out = EigenVector<T>::Flatten(*moment_out);
         auto ms_out = EigenVector<T>::Flatten(*mean_square_out);
-    auto& place = *ctx.template device_context<DeviceContext>().eigen_device();
-
-    Eigen::DSizes<int, 1> grad_dsize(static_cast<int>(grad->numel()));
 
         ms_out.device(place) = rho * ms + (1 - rho) * g * g;
         if (centered) {
-      auto mg = EigenVector<T>::Flatten(*ctx.Input<Tensor>("MeanGrad"));
-      auto* mean_grad_out = ctx.Output<Tensor>("MeanGradOut");
-      mean_grad_out->mutable_data<T>(ctx.GetPlace());
+          auto &mg_tensor = *ctx.Input<LoDTensor>("MeanGrad");
+          auto mg = EigenVector<T>::Flatten(mg_tensor);
+          auto *mean_grad_out = ctx.Output<LoDTensor>("MeanGradOut");
+          PADDLE_ENFORCE(&mg_tensor, mean_grad_out,
+                         "MeanGrad and MeanGradOut must be the same Tensor");
           auto mg_out = EigenVector<T>::Flatten(*mean_grad_out);
 
           mg_out.device(place) = rho * mg + (1 - rho) * g;
-      mom_out.device(place) = momentum * mom +
-                              lr.broadcast(grad_dsize) * g /
-                                  (ms_out - mg_out.square() + epsilon).sqrt();
-    } else {
           mom_out.device(place) =
               momentum * mom +
-          lr.broadcast(grad_dsize) * g / (ms_out + epsilon).sqrt();
+              lr_value * g / (ms_out - mg_out.square() + epsilon).sqrt();
+        } else {
+          mom_out.device(place) =
+              momentum * mom + lr_value * g / (ms_out + epsilon).sqrt();
         }
         p_out.device(place) = p - mom_out;
+      } else {
+        DenseRmspropGradFunctor<T> grad_func(grad_tensor.data<T>());
+        platform::ForRange<DeviceContext> for_range(dev_ctx, limit);
+        if (centered) {
+          auto &mg_tensor = *ctx.Input<LoDTensor>("MeanGrad");
+          auto *mean_grad_out = ctx.Output<LoDTensor>("MeanGradOut");
+          PADDLE_ENFORCE(&mg_tensor, mean_grad_out,
+                         "MeanGrad and MeanGradOut must be the same Tensor");
+          for_range(CenteredRmspropFunctor<T, DenseRmspropGradFunctor<T>>(
+              param_out->mutable_data<T>(ctx.GetPlace()),
+              mean_square_out->mutable_data<T>(ctx.GetPlace()),
+              moment_out->mutable_data<T>(ctx.GetPlace()),
+              mean_grad_out->mutable_data<T>(ctx.GetPlace()),
+              lr_tensor.data<T>(), rho, epsilon, momentum, grad_func));
+        } else {
+          for_range(UncenteredRmspropFunctor<T, DenseRmspropGradFunctor<T>>(
+              param_out->mutable_data<T>(ctx.GetPlace()),
+              mean_square_out->mutable_data<T>(ctx.GetPlace()),
+              moment_out->mutable_data<T>(ctx.GetPlace()), lr_tensor.data<T>(),
+              rho, epsilon, momentum, grad_func));
+        }
+      }
+    } else if (grad_var->IsType<framework::SelectedRows>()) {
+      auto &grad = grad_var->Get<framework::SelectedRows>();
+      auto *merged_grad = const_cast<framework::Scope &>(ctx.scope())
+                              .Var()
+                              ->GetMutable<framework::SelectedRows>();
+
+      math::scatter::MergeAdd<DeviceContext, T> merge_func;
+      merge_func(dev_ctx, grad, merged_grad);
+
+      platform::ForRange<DeviceContext> for_range(dev_ctx, limit);
+      const int64_t *rows;
+#ifdef PADDLE_WITH_CUDA
+      if (platform::is_gpu_place(ctx.GetPlace())) {
+        rows = merged_grad->rows().CUDAData(ctx.GetPlace());
+      } else {
+#endif
+        rows = merged_grad->rows().data();
+#ifdef PADDLE_WITH_CUDA
+      }
+#endif
+      auto &merged_tensor = merged_grad->value();
+      int64_t row_count = merged_grad->rows().size();
+      int64_t row_numel = merged_tensor.numel() / row_count;
+      SparseRmspropGradFunctor<T> grad_func(merged_tensor.data<T>(), rows,
+                                            row_numel, row_count);
+
+      if (centered) {
+        auto &mg_tensor = *ctx.Input<LoDTensor>("MeanGrad");
+        auto *mean_grad_out = ctx.Output<LoDTensor>("MeanGradOut");
+        PADDLE_ENFORCE(&mg_tensor, mean_grad_out,
+                       "MeanGrad and MeanGradOut must be the same Tensor");
+        for_range(CenteredRmspropFunctor<T, SparseRmspropGradFunctor<T>>(
+            param_out->mutable_data<T>(ctx.GetPlace()),
+            mean_square_out->mutable_data<T>(ctx.GetPlace()),
+            moment_out->mutable_data<T>(ctx.GetPlace()),
+            mean_grad_out->mutable_data<T>(ctx.GetPlace()), lr_tensor.data<T>(),
+            rho, epsilon, momentum, grad_func));
+      } else {
+        for_range(UncenteredRmspropFunctor<T, SparseRmspropGradFunctor<T>>(
+            param_out->mutable_data<T>(ctx.GetPlace()),
+            mean_square_out->mutable_data<T>(ctx.GetPlace()),
+            moment_out->mutable_data<T>(ctx.GetPlace()), lr_tensor.data<T>(),
+            rho, epsilon, momentum, grad_func));
+      }
+    } else {
+      PADDLE_THROW("RMSProp only supports LoDTensor or SelectedRows gradient");
+    }
   }
 };
 

python/paddle/fluid/tests/unittests/test_rmsprop_op.py

@@ -19,33 +19,76 @@ import unittest
 import numpy as np
 import paddle.fluid.core as core
 from paddle.fluid.op import Operator
+import paddle.fluid as fluid
+
+
+def create_selected_rows_and_tensor(scope, place, height, row_num,
+                                    embedding_size):
+    sr = scope.var("@selected_rows@").get_selected_rows()
+    tensor = scope.var("grad").get_tensor()
+
+    rows = np.random.random_integers(
+        low=0, high=height - 1, size=[row_num, ]).astype('int64')
+    sr_val = np.random.random(size=[row_num, embedding_size]).astype('float32')
+
+    sr.set_height(height)
+    sr.set_rows(rows)
+    sr.get_tensor().set(sr_val, place)
+
+    tensor_val = np.zeros(shape=[height, embedding_size], dtype='float32')
+    for i in range(row_num):
+        row = rows[i]
+        tensor_val[row, :] = tensor_val[row, :] + sr_val[i, :]
+
+    tensor.set(tensor_val, place)
+    return tensor_val, sr_val
 
 
 class TestBase(unittest.TestCase):
-    def setup(self, centered, epsilon=1e-6):
+    def setup(self,
+              place,
+              is_sparse,
+              centered,
+              size,
+              row_num=None,
+              epsilon=1e-6):
         np.random.seed(5)  # fix seed
 
+        self.scope = fluid.global_scope()
+        self.place = place
+
         self.param_name = "param"
-        self.param = np.random.random((123, 321)).astype("float32")
+        self.param = np.random.random(size).astype("float32")
 
         self.mean_square_name = "mean_square"
-        self.mean_square = np.random.random((123, 321)).astype("float32")
+        self.mean_square = np.random.uniform(
+            low=1, high=2, size=size).astype("float32")
 
         self.mean_grad_name = "mean_grad"
-        self.mean_grad = np.random.random((123, 321)).astype("float32")
+        self.mean_grad = np.random.random(size).astype("float32")
 
         self.lr_name = "lr"
         self.learning_rate = np.array([0.01]).astype("float32")
 
         self.grad_name = "grad"
-        self.grad = np.random.random((123, 321)).astype("float32")
+
+        self.is_sparse = is_sparse
+        if self.is_sparse:
+            self.grad_sr_name = "@selected_rows@"
+            self.grad, self.grad_sr = create_selected_rows_and_tensor(
+                self.scope, place, size[0], row_num, size[1])
+        else:
+            self.grad = np.random.random(size).astype("float32")
+            grad_tensor = self.scope.var(self.grad_name).get_tensor()
+            grad_tensor.set(self.grad, place)
 
         self.moment_name = "moment"
-        self.moment = np.zeros((123, 321)).astype("float32")
+        self.moment = np.random.uniform(
+            low=0, high=1, size=size).astype("float32")
 
         self.epsilon = epsilon
         self.decay = 0.9
-        self.momentum = 0.0
+        self.momentum = 0.1
         self.centered = centered
 
         self.ms_out = self.decay * self.mean_square + (1 - self.decay
@@ -61,118 +104,122 @@ class TestBase(unittest.TestCase):
 
         self.param_out = self.param - self.moment_out
 
-    def check(self,
-              actual_t,
-              expect_t,
-              place,
-              out_name,
-              atol=1e-5,
-              equal_nan=False):
-        self.assertTrue(
-            np.allclose(
-                actual_t, expect_t, atol=atol, equal_nan=equal_nan),
-            "Output (" + out_name + ") has diff at " + str(place) + "\nExpect "
-            + str(expect_t) + "\n" + "But Got" + str(actual_t))
-
-
-class TestRmspropOp(TestBase):
-    def check_with_place(self, place, centered, epsilon):
-        self.setup(centered, epsilon)
-        scope = core.Scope()
-
         # create and initialize Param Variable
-        param = scope.var(self.param_name).get_tensor()
-        param.set(self.param, place)
+        self.param_tensor = self.scope.var(self.param_name).get_tensor()
+        self.param_tensor.set(self.param, place)
 
-        mean_square = scope.var(self.mean_square_name).get_tensor()
-        mean_square.set(self.mean_square, place)
+        self.mean_square_tensor = self.scope.var(
+            self.mean_square_name).get_tensor()
+        self.mean_square_tensor.set(self.mean_square, place)
 
-        lr = scope.var(self.lr_name).get_tensor()
+        lr = self.scope.var(self.lr_name).get_tensor()
         lr.set(self.learning_rate, place)
 
-        grad = scope.var(self.grad_name).get_tensor()
-        grad.set(self.grad, place)
+        self.moment_tensor = self.scope.var(self.moment_name).get_tensor()
+        self.moment_tensor.set(self.moment, place)
 
-        moment = scope.var(self.moment_name).get_tensor()
-        moment.set(self.moment, place)
+        if self.centered:
+            self.mean_grad_tensor = self.scope.var(
+                self.mean_grad_name).get_tensor()
+            self.mean_grad_tensor.set(self.mean_grad, place)
+
+    def check(self, actual_t, expect_t, place, out_name, atol=1e-5):
+        self.assertTrue(
+            np.allclose(
+                actual_t, expect_t, atol=atol),
+            "Output (" + out_name + ") has diff at " + str(place) + "\nExpect "
+            + str(expect_t) + "\n" + "But Got" + str(actual_t))
 
-        # create and run sgd operator
 
-        if self.centered:
-            mean_grad = scope.var(self.mean_grad_name).get_tensor()
-            mean_grad.set(self.mean_grad, place)
-
-            rmsprop_op = Operator(
-                "rmsprop",
-                Param=self.param_name,
-                Grad=self.grad_name,
-                MeanSquare=self.mean_square_name,
-                MeanGrad=self.mean_grad_name,
-                Moment=self.moment_name,
-                LearningRate=self.lr_name,
-                ParamOut=self.param_name,
-                MeanSquareOut=self.mean_square_name,
-                MomentOut=self.moment_name,
-                MeanGradOut=self.mean_grad_name,
-                epsilon=self.epsilon,
-                decay=self.decay,
-                momentum=self.momentum,
-                centered=True)
-        else:
-            rmsprop_op = Operator(
-                "rmsprop",
-                Param=self.param_name,
-                Grad=self.grad_name,
-                MeanSquare=self.mean_square_name,
-                Moment=self.moment_name,
-                LearningRate=self.lr_name,
-                ParamOut=self.param_name,
-                MeanSquareOut=self.mean_square_name,
-                MomentOut=self.moment_name,
-                epsilon=self.epsilon,
-                decay=self.decay,
-                momentum=self.momentum,
-                centered=False)
-
-        rmsprop_op.run(scope, place)
-
-        atol = 1e-5
-        equal_nan = False
+class TestRmspropOp(TestBase):
+    def check_with_place(self,
+                         place,
+                         is_sparse,
+                         centered,
+                         size,
+                         row_num=None,
+                         epsilon=1e-6):
+        self.setup(place, is_sparse, centered, size, row_num, epsilon)
+        self.run_and_check()
+
+    def run_and_check(self):
+        grad_name = self.grad_sr_name if self.is_sparse else self.grad_name
+
+        kwargs = {
+            'Param': self.param_name,
+            'Grad': grad_name,
+            'MeanSquare': self.mean_square_name,
+            'Moment': self.moment_name,
+            'LearningRate': self.lr_name,
+            'ParamOut': self.param_name,
+            'MeanSquareOut': self.mean_square_name,
+            'MomentOut': self.moment_name,
+            'epsilon': self.epsilon,
+            'decay': self.decay,
+            'momentum': self.momentum,
+            'centered': self.centered
+        }
 
         if self.centered:
-            atol = 1e-3
-            equal_nan = True
+            kwargs['MeanGrad'] = self.mean_grad_name
+            kwargs['MeanGradOut'] = self.mean_grad_name
+
+        rmsprop_op = Operator('rmsprop', **kwargs)
+        atol = 1e-6
+
+        rmsprop_op.run(self.scope, self.place)
 
         self.check(
-            np.array(mean_square), self.ms_out, place, self.mean_square_name)
+            np.array(self.mean_square_tensor),
+            self.ms_out,
+            self.place,
+            self.mean_square_name,
+            atol=atol)
         self.check(
-            np.array(moment),
+            np.array(self.moment_tensor),
             self.moment_out,
-            place,
+            self.place,
             self.moment_name,
-            atol=atol,
-            equal_nan=equal_nan)
+            atol=atol)
         self.check(
-            np.array(param),
+            np.array(self.param_tensor),
             self.param_out,
-            place,
+            self.place,
             self.param_name,
-            atol=atol,
-            equal_nan=equal_nan)
+            atol=atol)
 
         if self.centered:
             self.check(
-                np.array(mean_grad), self.mg_out, place, self.mean_grad_name)
+                np.array(self.mean_grad_tensor), self.mg_out, self.place,
+                self.mean_grad_name)
 
     def test_rmsprop(self):
         places = [core.CPUPlace()]
         if core.is_compiled_with_cuda():
             places.append(core.CUDAPlace(0))
+
+        size = (128, 320)
         for place in places:
-            self.check_with_place(place, False, 1e-6)
-            self.check_with_place(place, False, 1e-10)
-            self.check_with_place(place, True, 1e-6)
-            self.check_with_place(place, True, 1e-10)
+            for centered in [False, True]:
+                with fluid.scope_guard(core.Scope()):
+                    self.check_with_place(
+                        place, is_sparse=False, centered=centered, size=size)
+
+                with fluid.scope_guard(core.Scope()):
+                    self.check_with_place(
+                        place,
+                        is_sparse=True,
+                        centered=centered,
+                        row_num=512,
+                        size=size)
+
+                with fluid.scope_guard(core.Scope()):
+                    self.check_with_place(
+                        place,
+                        is_sparse=True,
+                        centered=centered,
+                        row_num=60,
+                        size=size)
 
 
 if __name__ == "__main__":