activation_functions.h 4.7 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
/* Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserve.

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 <math.h>
#include "paddle/platform/hostdevice.h"
18
#include "paddle/platform/enforce.h"
19 20 21 22 23 24 25 26 27 28 29 30 31 32

#ifdef __AVX__
#include <immintrin.h>
#endif

namespace paddle {
namespace operators {
namespace math {
namespace detail {

#define SIGMOID_THRESHOLD_MIN -40.0
#define SIGMOID_THRESHOLD_MAX 13.0
#define EXP_MAX_INPUT 40.0

33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53
enum ActivationType {
  kSigmoid,
  kReLU,
  kTanh,
  kIdentity,
};

inline ActivationType GetActivationType (const std::string &type) {
  if (type == "sigmoid") {
    return ActivationType::kSigmoid;
  } else if (type == "relu") {
    return ActivationType::kReLU;
  } else if (type == "tanh") {
    return ActivationType::kTanh;
  } else if (type == "identity") {
    return ActivationType::kIdentity;
  }
  PADDLE_THROW("Not support type %s.", type);
}


54 55 56
namespace forward {

template <typename T>
57
DEVICE T Identity(const T a) {
58 59 60 61
  return a;
}

template <typename T>
62
DEVICE T Relu(const T a) {
63 64 65 66
  return a > static_cast<T>(0.0) ? a : static_cast<T>(0.0);
}

template <typename T>
67
DEVICE T Sigmoid(const T a) {
68 69 70 71 72 73 74
  const T min = SIGMOID_THRESHOLD_MIN;
  const T max = SIGMOID_THRESHOLD_MAX;
  T tmp = (a < min) ? min : ((a > max) ? max : a);
  return static_cast<T>(1.0) / (static_cast<T>(1.0) + exp(-tmp));
}

template <typename T>
75
DEVICE T Tanh(const T a) {
76 77 78 79 80 81 82 83 84 85
  T tmp = -2.0 * a;
  tmp = (tmp > EXP_MAX_INPUT) ? EXP_MAX_INPUT : tmp;
  return (2.0 / (1.0 + exp(tmp))) - 1.0;
}

}  // namespace forward

namespace backward {

template <typename T>
86
DEVICE T Identity(const T a, const T b) {
87 88 89 90
  return a;
}

template <typename T>
91
DEVICE T Relu(const T a, const T b) {
92 93 94 95
  return a * (b > 0.0 ? 1.0 : 0.0);
}

template <typename T>
96
DEVICE T Sigmoid(const T a, const T b) {
97 98 99 100
  return a * b * (1.0 - b);
}

template <typename T>
101
DEVICE T Tanh(const T a, const T b) {
102 103 104 105 106 107 108 109 110 111 112 113
  return a * (1.0 - b * b);
}

}  // namespace backward

template <typename T>
struct Active {
  typedef T (*Act)(T);
  typedef T (*ActGrad)(T, T);
};

static DEVICE Active<float>::Act kActFloat[] = {
114 115
    &forward::Sigmoid<float>, &forward::Relu<float>, &forward::Tanh<float>,
    &forward::Identity<float>};
116 117

static DEVICE Active<float>::ActGrad kActGradFloat[] = {
118 119
    &backward::Sigmoid<float>, &backward::Relu<float>, &backward::Tanh<float>,
    &backward::Identity<float>};
120 121

static DEVICE Active<double>::Act kActDouble[] = {
122 123
    &forward::Sigmoid<double>, &forward::Relu<double>, &forward::Tanh<double>,
    &forward::Identity<double>};
124 125

static DEVICE Active<double>::ActGrad kActGradDouble[] = {
126 127
    &backward::Sigmoid<double>, &backward::Relu<double>,
    &backward::Tanh<double>, &backward::Identity<double>};
128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152

namespace forward {
inline DEVICE float activation(float a, int index) {
  return kActFloat[index](a);
}

inline DEVICE double activation(double a, int index) {
  return kActDouble[index](a);
}

}  // namespace forward

namespace backward {
inline DEVICE float activation(float a, float b, int index) {
  return kActGradFloat[index](a, b);
}

inline DEVICE double activation(double a, double b, int index) {
  return kActGradDouble[index](a, b);
}
}  // namespace backward

#ifdef __AVX__
namespace forward {
namespace avx {
153 154 155 156
__m256 Relu(const __m256 a);
__m256 Sigmoid(const __m256 a);
__m256 Tanh(const __m256 a);
__m256 Identity(const __m256 a);
157 158 159 160 161
}  // namespace avx
}  // namespace forward

namespace backward {
namespace avx {
162 163 164 165
__m256 Relu(const __m256 a, const __m256 b);
__m256 Sigmoid(const __m256 a, const __m256 b);
__m256 Tanh(const __m256 a, const __m256 b);
__m256 Identity(const __m256 a, const __m256 b);
166 167 168 169
}  // namespace avx
}  // namespace backward

static Active<__m256>::Act kActAvx[] = {
170 171
    &forward::avx::Sigmoid, &forward::avx::Relu, &forward::avx::Tanh,
    &forward::avx::Identity};
172 173

static Active<__m256>::ActGrad kActGradAvx[] = {
174 175
    &backward::avx::Sigmoid, &backward::avx::Relu, &backward::avx::Tanh,
    &backward::avx::Identity};
176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192

namespace forward {
inline __m256 activation(__m256 a, int index) { return kActAvx[index](a); }
}  // namespace forward

namespace backward {
inline __m256 activation(__m256 a, __m256 b, int index) {
  return kActGradAvx[index](a, b);
}
}  // namespace backward

#endif

}  // namespace detail
}  // namespace math
}  // namespace operators
}  // namespace paddle