-
Notifications
You must be signed in to change notification settings - Fork 5
/
oplu.cu
77 lines (61 loc) · 2.03 KB
/
oplu.cu
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
#include <algorithm>
#include <vector>
#include "caffe/layers/oplu_layer.hpp"
namespace caffe {
template <typename Dtype>
__global__ void OPLUForward(const int n, const Dtype* in, Dtype* out) {
CUDA_KERNEL_LOOP2(index, n) {
if (in[index] > in[index+1]) {
out[index] = in[index];
out[index+1] = in[index+1];
}
else {
out[index+1] = in[index];
out[index] = in[index+1];
}
}
}
template <typename Dtype>
void OPLULayer<Dtype>::Forward_gpu(const vector<Blob<Dtype>*>& bottom,
const 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)
OPLUForward<Dtype><<<CAFFE_GET_BLOCKS(count), CAFFE_CUDA_NUM_THREADS>>>(
count, bottom_data, top_data);
CUDA_POST_KERNEL_CHECK;
}
template <typename Dtype>
__global__ void OPLUBackward(const int n, const Dtype* in_diff,
const Dtype* out_data, const Dtype* in_data,
Dtype* out_diff) {
CUDA_KERNEL_LOOP2(index, n) {
if (in_data[index] > in_data[index+1]) {
out_diff[index] = in_diff[index];
out_diff[index+1] = in_diff[index+1];
}
else {
out_diff[index+1] = in_diff[index];
out_diff[index] = in_diff[index+1];
}
}
}
template <typename Dtype>
void OPLULayer<Dtype>::Backward_gpu(const vector<Blob<Dtype>*>& top,
const vector<bool>& propagate_down,
const vector<Blob<Dtype>*>& bottom) {
if (propagate_down[0]) {
const Dtype* bottom_data = bottom[0]->gpu_data();
const Dtype* top_diff = top[0]->gpu_diff();
const Dtype* top_data = top[0]->gpu_data();
Dtype* bottom_diff = bottom[0]->mutable_gpu_diff();
const int count = bottom[0]->count();
// NOLINT_NEXT_LINE(whitespace/operators)
OPLUBackward<Dtype><<<CAFFE_GET_BLOCKS(count), CAFFE_CUDA_NUM_THREADS>>>(
count, top_diff, top_data, bottom_data, bottom_diff);
CUDA_POST_KERNEL_CHECK;
}
}
INSTANTIATE_LAYER_GPU_FUNCS(OPLULayer);
} // namespace caffe