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
|
// Copyright 2014 Tobias Domhan
#include <algorithm>
#include <cmath>
#include <vector>
#include "caffe/layer.hpp"
#include "caffe/vision_layers.hpp"
using std::max;
namespace caffe {
template <typename Dtype>
__device__ inline Dtype sigmoid_gpu(Dtype x) {
return 1. / (1. + exp(-x));
}
template <typename Dtype>
__global__ void SigmoidForward(const int n, const Dtype* in, Dtype* out) {
int index = threadIdx.x + blockIdx.x * blockDim.x;
if (index < n) {
out[index] = sigmoid_gpu(in[index]);
}
}
template <typename Dtype>
void SigmoidLayer<Dtype>::Forward_gpu(const vector<Blob<Dtype>*>& bottom,
vector<Blob<Dtype>*>* top) {
const Dtype* bottom_data = bottom[0]->gpu_data();
Dtype* top_data = (*top)[0]->mutable_gpu_data();
const int count = bottom[0]->count();
// NOLINT_NEXT_LINE(whitespace/operators)
SigmoidForward<Dtype><<<CAFFE_GET_BLOCKS(count), CAFFE_CUDA_NUM_THREADS>>>(
count, bottom_data, top_data);
CUDA_POST_KERNEL_CHECK;
// << " count: " << count << " bottom_data: "
// << (unsigned long)bottom_data
// << " top_data: " << (unsigned long)top_data
// << " blocks: " << CAFFE_GET_BLOCKS(count)
// << " threads: " << CAFFE_CUDA_NUM_THREADS;
}
template <typename Dtype>
__global__ void SigmoidBackward(const int n, const Dtype* in_diff,
const Dtype* in_data, Dtype* out_diff) {
int index = threadIdx.x + blockIdx.x * blockDim.x;
if (index < n) {
Dtype sigmoid_x = sigmoid_gpu(in_data[index]);
out_diff[index] = in_diff[index] * sigmoid_x * (1 - sigmoid_x);
}
}
template <typename Dtype>
Dtype SigmoidLayer<Dtype>::Backward_gpu(const vector<Blob<Dtype>*>& top,
const bool propagate_down,
vector<Blob<Dtype>*>* bottom) {
if (propagate_down) {
const Dtype* bottom_data = (*bottom)[0]->gpu_data();
const Dtype* top_diff = top[0]->gpu_diff();
Dtype* bottom_diff = (*bottom)[0]->mutable_gpu_diff();
const int count = (*bottom)[0]->count();
// NOLINT_NEXT_LINE(whitespace/operators)
SigmoidBackward<Dtype><<<CAFFE_GET_BLOCKS(count), CAFFE_CUDA_NUM_THREADS>>>(
count, top_diff, bottom_data, bottom_diff);
CUDA_POST_KERNEL_CHECK;
}
return Dtype(0);
}
INSTANTIATE_CLASS(SigmoidLayer);
} // namespace caffe
|
