1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
|
/* Copyright 2015 The TensorFlow 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.
==============================================================================*/
// See docs in ../ops/nn_ops.cc.
#define EIGEN_USE_THREADS
#include "tensorflow/core/kernels/softsign_op.h"
#include "third_party/eigen3/unsupported/Eigen/CXX11/Tensor"
#include "tensorflow/core/framework/numeric_op.h"
#include "tensorflow/core/framework/op_kernel.h"
#include "tensorflow/core/framework/register_types.h"
#include "tensorflow/core/framework/tensor.h"
#include "tensorflow/core/lib/core/errors.h"
namespace tensorflow {
typedef Eigen::ThreadPoolDevice CPUDevice;
typedef Eigen::GpuDevice GPUDevice;
template <typename Device, typename T>
class SoftsignOp : public UnaryElementWiseOp<T, SoftsignOp<Device, T>> {
public:
explicit SoftsignOp(OpKernelConstruction* context)
: UnaryElementWiseOp<T, SoftsignOp<Device, T>>(context) {}
void Operate(OpKernelContext* context, const Tensor& input, Tensor* output) {
functor::Softsign<Device, T> functor;
functor(context->eigen_device<Device>(), input.flat<T>(),
output->flat<T>());
}
};
template <typename Device, typename T>
class SoftsignGradOp
: public BinaryElementWiseOp<T, SoftsignGradOp<Device, T>> {
public:
explicit SoftsignGradOp(OpKernelConstruction* context)
: BinaryElementWiseOp<T, SoftsignGradOp<Device, T>>(context) {}
void OperateNoTemplate(OpKernelContext* context, const Tensor& g,
const Tensor& a, Tensor* output);
// INPUTS:
// g (gradients): backpropagated gradients
// a (inputs): inputs that were passed to SoftsignOp()
// OUTPUT:
// gradients to backprop
template <int NDIMS>
void Operate(OpKernelContext* context, const Tensor& g, const Tensor& a,
Tensor* output) {
OperateNoTemplate(context, g, a, output);
}
};
template <typename Device, typename T>
void SoftsignGradOp<Device, T>::OperateNoTemplate(OpKernelContext* context,
const Tensor& g,
const Tensor& a,
Tensor* output) {
OP_REQUIRES(context, a.IsSameSize(g),
errors::InvalidArgument("g and a must be the same size"));
functor::SoftsignGrad<Device, T> functor;
functor(context->eigen_device<Device>(), g.flat<T>(), a.flat<T>(),
output->flat<T>());
}
#define REGISTER_KERNELS(type) \
REGISTER_KERNEL_BUILDER( \
Name("Softsign").Device(DEVICE_CPU).TypeConstraint<type>("T"), \
SoftsignOp<CPUDevice, type>); \
REGISTER_KERNEL_BUILDER( \
Name("SoftsignGrad").Device(DEVICE_CPU).TypeConstraint<type>("T"), \
SoftsignGradOp<CPUDevice, type>);
TF_CALL_FLOAT_TYPES(REGISTER_KERNELS);
#undef REGISTER_KERNELS
#if GOOGLE_CUDA
// Forward declarations of the functor specializations for GPU.
namespace functor {
#define DECLARE_GPU_SPEC(T) \
template <> \
void Softsign<GPUDevice, T>::operator()( \
const GPUDevice& d, typename TTypes<T>::ConstTensor features, \
typename TTypes<T>::Tensor activations); \
extern template struct Softsign<GPUDevice, T>; \
\
template <> \
void SoftsignGrad<GPUDevice, T>::operator()( \
const GPUDevice& d, typename TTypes<T>::ConstTensor gradients, \
typename TTypes<T>::ConstTensor features, \
typename TTypes<T>::Tensor backprops); \
extern template struct SoftsignGrad<GPUDevice, T>;
TF_CALL_GPU_NUMBER_TYPES(DECLARE_GPU_SPEC);
} // namespace functor
// Registration of the GPU implementations.
#define REGISTER_GPU_KERNELS(type) \
REGISTER_KERNEL_BUILDER( \
Name("Softsign").Device(DEVICE_GPU).TypeConstraint<type>("T"), \
SoftsignOp<GPUDevice, type>); \
REGISTER_KERNEL_BUILDER( \
Name("SoftsignGrad").Device(DEVICE_GPU).TypeConstraint<type>("T"), \
SoftsignGradOp<GPUDevice, type>);
TF_CALL_GPU_NUMBER_TYPES(REGISTER_GPU_KERNELS);
#undef REGISTER_GPU_KERNELS
#endif // GOOGLE_CUDA
} // namespace tensorflow
|
