softmax_op.h 2.4 KB
Newer Older
1 2
/* Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserve.

Q
Qiao Longfei 已提交
3 4 5
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
6

Q
Qiao Longfei 已提交
7
    http://www.apache.org/licenses/LICENSE-2.0
8

Q
Qiao Longfei 已提交
9 10 11 12 13
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. */
14 15

#pragma once
D
dongzhihong 已提交
16 17
#include "paddle/framework/eigen.h"
#include "paddle/framework/op_registry.h"
C
caoying03 已提交
18
#include "paddle/operators/math/softmax.h"
19 20 21 22

namespace paddle {
namespace operators {

D
dongzhihong 已提交
23 24 25 26 27
using Tensor = framework::Tensor;
template <typename T, int MajorType = Eigen::RowMajor,
          typename IndexType = Eigen::DenseIndex>
using EigenMatrix = framework::EigenMatrix<T, MajorType, IndexType>;

Q
qijun 已提交
28
template <typename Place, typename T>
D
dongzhihong 已提交
29
class SoftmaxKernel : public framework::OpKernel {
30
 public:
D
dongzhihong 已提交
31
  void Compute(const framework::ExecutionContext& context) const override {
32 33
    auto X = context.Input<Tensor>("X");
    auto Y = context.Output<Tensor>("Y");
Q
qijun 已提交
34

C
caoying03 已提交
35 36
    // allocate memory on device.
    Y->mutable_data<T>(context.GetPlace());
Q
qijun 已提交
37

38
    math::SoftmaxFunctor<Place, T>()(context, X, Y);
39 40
  }
};
Q
Qiao Longfei 已提交
41 42

template <typename Place, typename T>
D
dongzhihong 已提交
43
class SoftmaxGradKernel : public framework::OpKernel {
44
 public:
D
dongzhihong 已提交
45
  void Compute(const framework::ExecutionContext& context) const override {
46 47 48
    auto Y = context.Input<Tensor>("Y");
    auto dY = context.Input<Tensor>(framework::GradVarName("Y"));
    auto dX = context.Output<Tensor>(framework::GradVarName("X"));
Q
Qiao Longfei 已提交
49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71
    dX->mutable_data<T>(context.GetPlace());

    const int batch_size = Y->dims()[0];
    const int class_num = Y->dims()[1];

    Eigen::DSizes<int, 1> along_class(1);
    Eigen::DSizes<int, 2> batch_by_one(batch_size, 1);
    Eigen::DSizes<int, 2> one_by_class(1, class_num);

    auto Y_eigen = EigenMatrix<T>::From(*Y);
    auto dY_eigen = EigenMatrix<T>::From(*dY);
    auto dX_eigen = EigenMatrix<T>::From(*dX);
    auto place = context.GetEigenDevice<Place>();

    auto dot = (Y_eigen * dY_eigen)
                   .sum(along_class)
                   .eval()
                   .reshape(batch_by_one)
                   .broadcast(one_by_class);
    dX_eigen.device(place) = (dY_eigen - dot) * Y_eigen;
  }
};

72 73
}  // namespace operators
}  // namespace paddle