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
126
127
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
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
|
// Copyright 2013 Yangqing Jia
#include <algorithm>
#include <cmath>
#include <cfloat>
#include <vector>
#include "caffe/layer.hpp"
#include "caffe/vision_layers.hpp"
#include "caffe/util/math_functions.hpp"
#include "caffe/util/io.hpp"
using std::max;
namespace caffe {
const float kLOG_THRESHOLD = 1e-20;
template <typename Dtype>
void MultinomialLogisticLossLayer<Dtype>::SetUp(
const vector<Blob<Dtype>*>& bottom, vector<Blob<Dtype>*>* top) {
CHECK_EQ(bottom.size(), 2) << "Loss Layer takes two blobs as input.";
CHECK_EQ(top->size(), 0) << "Loss Layer takes no output.";
CHECK_EQ(bottom[0]->num(), bottom[1]->num())
<< "The data and label should have the same number.";
CHECK_EQ(bottom[1]->channels(), 1);
CHECK_EQ(bottom[1]->height(), 1);
CHECK_EQ(bottom[1]->width(), 1);
}
template <typename Dtype>
Dtype MultinomialLogisticLossLayer<Dtype>::Forward_cpu(
const vector<Blob<Dtype>*>& bottom, vector<Blob<Dtype>*>* top) {
const Dtype* bottom_data = bottom[0]->cpu_data();
const Dtype* bottom_label = bottom[1]->cpu_data();
int num = bottom[0]->num();
int dim = bottom[0]->count() / bottom[0]->num();
Dtype loss = 0;
for (int i = 0; i < num; ++i) {
int label = static_cast<int>(bottom_label[i]);
Dtype prob = max(bottom_data[i * dim + label], Dtype(kLOG_THRESHOLD));
loss -= log(prob);
}
return loss / num;
}
template <typename Dtype>
void MultinomialLogisticLossLayer<Dtype>::Backward_cpu(
const vector<Blob<Dtype>*>& top, const bool propagate_down,
vector<Blob<Dtype>*>* bottom) {
const Dtype* bottom_data = (*bottom)[0]->cpu_data();
const Dtype* bottom_label = (*bottom)[1]->cpu_data();
Dtype* bottom_diff = (*bottom)[0]->mutable_cpu_diff();
int num = (*bottom)[0]->num();
int dim = (*bottom)[0]->count() / (*bottom)[0]->num();
memset(bottom_diff, 0, sizeof(Dtype) * (*bottom)[0]->count());
for (int i = 0; i < num; ++i) {
int label = static_cast<int>(bottom_label[i]);
Dtype prob = max(bottom_data[i * dim + label], Dtype(kLOG_THRESHOLD));
bottom_diff[i * dim + label] = -1. / prob / num;
}
}
template <typename Dtype>
void InfogainLossLayer<Dtype>::SetUp(
const vector<Blob<Dtype>*>& bottom, vector<Blob<Dtype>*>* top) {
CHECK_EQ(bottom.size(), 2) << "Loss Layer takes two blobs as input.";
CHECK_EQ(top->size(), 0) << "Loss Layer takes no output.";
CHECK_EQ(bottom[0]->num(), bottom[1]->num())
<< "The data and label should have the same number.";
CHECK_EQ(bottom[1]->channels(), 1);
CHECK_EQ(bottom[1]->height(), 1);
CHECK_EQ(bottom[1]->width(), 1);
BlobProto blob_proto;
ReadProtoFromBinaryFile(this->layer_param_.source(), &blob_proto);
infogain_.FromProto(blob_proto);
CHECK_EQ(infogain_.num(), 1);
CHECK_EQ(infogain_.channels(), 1);
CHECK_EQ(infogain_.height(), infogain_.width());
}
template <typename Dtype>
Dtype InfogainLossLayer<Dtype>::Forward_cpu(const vector<Blob<Dtype>*>& bottom,
vector<Blob<Dtype>*>* top) {
const Dtype* bottom_data = bottom[0]->cpu_data();
const Dtype* bottom_label = bottom[1]->cpu_data();
const Dtype* infogain_mat = infogain_.cpu_data();
int num = bottom[0]->num();
int dim = bottom[0]->count() / bottom[0]->num();
CHECK_EQ(infogain_.height(), dim);
Dtype loss = 0;
for (int i = 0; i < num; ++i) {
int label = static_cast<int>(bottom_label[i]);
for (int j = 0; j < dim; ++j) {
Dtype prob = max(bottom_data[i * dim + j], Dtype(kLOG_THRESHOLD));
loss -= infogain_mat[label * dim + j] * log(prob);
}
}
return loss / num;
}
template <typename Dtype>
void InfogainLossLayer<Dtype>::Backward_cpu(const vector<Blob<Dtype>*>& top,
const bool propagate_down,
vector<Blob<Dtype>*>* bottom) {
const Dtype* bottom_data = (*bottom)[0]->cpu_data();
const Dtype* bottom_label = (*bottom)[1]->cpu_data();
const Dtype* infogain_mat = infogain_.cpu_data();
Dtype* bottom_diff = (*bottom)[0]->mutable_cpu_diff();
int num = (*bottom)[0]->num();
int dim = (*bottom)[0]->count() / (*bottom)[0]->num();
CHECK_EQ(infogain_.height(), dim);
for (int i = 0; i < num; ++i) {
int label = static_cast<int>(bottom_label[i]);
for (int j = 0; j < dim; ++j) {
Dtype prob = max(bottom_data[i * dim + j], Dtype(kLOG_THRESHOLD));
bottom_diff[i * dim + j] = - infogain_mat[label * dim + j] / prob / num;
}
}
}
template <typename Dtype>
void EuclideanLossLayer<Dtype>::SetUp(
const vector<Blob<Dtype>*>& bottom, vector<Blob<Dtype>*>* top) {
CHECK_EQ(bottom.size(), 2) << "Loss Layer takes two blobs as input.";
CHECK_EQ(top->size(), 0) << "Loss Layer takes no as output.";
CHECK_EQ(bottom[0]->num(), bottom[1]->num())
<< "The data and label should have the same number.";
CHECK_EQ(bottom[0]->channels(), bottom[1]->channels());
CHECK_EQ(bottom[0]->height(), bottom[1]->height());
CHECK_EQ(bottom[0]->width(), bottom[1]->width());
difference_.Reshape(bottom[0]->num(), bottom[0]->channels(),
bottom[0]->height(), bottom[0]->width());
}
template <typename Dtype>
Dtype EuclideanLossLayer<Dtype>::Forward_cpu(const vector<Blob<Dtype>*>& bottom,
vector<Blob<Dtype>*>* top) {
int count = bottom[0]->count();
int num = bottom[0]->num();
caffe_sub(count, bottom[0]->cpu_data(), bottom[1]->cpu_data(),
difference_.mutable_cpu_data());
Dtype loss = caffe_cpu_dot(
count, difference_.cpu_data(), difference_.cpu_data()) / num / Dtype(2);
return loss;
}
template <typename Dtype>
void EuclideanLossLayer<Dtype>::Backward_cpu(const vector<Blob<Dtype>*>& top,
const bool propagate_down, vector<Blob<Dtype>*>* bottom) {
int count = (*bottom)[0]->count();
int num = (*bottom)[0]->num();
// Compute the gradient
caffe_cpu_axpby(count, Dtype(1) / num, difference_.cpu_data(), Dtype(0),
(*bottom)[0]->mutable_cpu_diff());
}
template <typename Dtype>
void AccuracyLayer<Dtype>::SetUp(
const vector<Blob<Dtype>*>& bottom, vector<Blob<Dtype>*>* top) {
CHECK_EQ(bottom.size(), 2) << "Accuracy Layer takes two blobs as input.";
CHECK_EQ(top->size(), 1) << "Accuracy Layer takes 1 output.";
CHECK_EQ(bottom[0]->num(), bottom[1]->num())
<< "The data and label should have the same number.";
CHECK_EQ(bottom[1]->channels(), 1);
CHECK_EQ(bottom[1]->height(), 1);
CHECK_EQ(bottom[1]->width(), 1);
(*top)[0]->Reshape(1, 2, 1, 1);
}
template <typename Dtype>
Dtype AccuracyLayer<Dtype>::Forward_cpu(const vector<Blob<Dtype>*>& bottom,
vector<Blob<Dtype>*>* top) {
Dtype accuracy = 0;
Dtype logprob = 0;
const Dtype* bottom_data = bottom[0]->cpu_data();
const Dtype* bottom_label = bottom[1]->cpu_data();
int num = bottom[0]->num();
int dim = bottom[0]->count() / bottom[0]->num();
for (int i = 0; i < num; ++i) {
// Accuracy
Dtype maxval = -FLT_MAX;
int max_id = 0;
for (int j = 0; j < dim; ++j) {
if (bottom_data[i * dim + j] > maxval) {
maxval = bottom_data[i * dim + j];
max_id = j;
}
}
if (max_id == static_cast<int>(bottom_label[i])) {
++accuracy;
}
Dtype prob = max(bottom_data[i * dim + static_cast<int>(bottom_label[i])],
Dtype(kLOG_THRESHOLD));
logprob -= log(prob);
}
// LOG(INFO) << "Accuracy: " << accuracy;
(*top)[0]->mutable_cpu_data()[0] = accuracy / num;
(*top)[0]->mutable_cpu_data()[1] = logprob / num;
// Accuracy layer should not be used as a loss function.
return Dtype(0);
}
INSTANTIATE_CLASS(MultinomialLogisticLossLayer);
INSTANTIATE_CLASS(InfogainLossLayer);
INSTANTIATE_CLASS(EuclideanLossLayer);
INSTANTIATE_CLASS(AccuracyLayer);
} // namespace caffe
|
