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
|
#include "caffe2/operators/tanh_op.h"
#include "caffe2/utils/eigen_utils.h"
#include <algorithm>
#include <functional>
#include <string>
#include <vector>
namespace caffe2 {
template <>
template <>
bool TanhGradientFunctor<CPUContext>::Forward<float>(
const std::vector<int>& Y_dims,
const std::vector<int>& /* dY_dims */,
const float* Y,
const float* dY,
float* dX,
CPUContext* /* context */) const {
const int size = std::accumulate(
// NOLINTNEXTLINE(modernize-use-transparent-functors)
Y_dims.cbegin(), Y_dims.cend(), 1, std::multiplies<int>());
ConstEigenVectorArrayMap<float> dY_arr(dY, size);
ConstEigenVectorArrayMap<float> Y_arr(Y, size);
EigenVectorMap<float>(dX, size) = dY_arr * (1 - Y_arr * Y_arr);
return true;
}
REGISTER_CPU_OPERATOR(
TanhGradient,
BinaryElementwiseOp<
TensorTypes<float>,
CPUContext,
TanhGradientFunctor<CPUContext>>);
namespace {
class GetTanhGradient : public GradientMakerBase {
using GradientMakerBase::GradientMakerBase;
std::vector<OperatorDef> GetGradientDefs() override {
return SingleGradientDef(
"TanhGradient",
"",
std::vector<std::string>{O(0), GO(0)},
std::vector<std::string>{GI(0)});
}
};
} // namespace
REGISTER_GRADIENT(Tanh, GetTanhGradient);
} // namespace caffe2
|
