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
|
/*********************************************************************
MLDemos: A User-Friendly visualization toolkit for machine learning
Copyright (C) 2010 Basilio Noris
Contact: mldemos@b4silio.com
This library is free software; you can redistribute it and/or
modify it under the terms of the GNU Lesser General Public
License as published by the Free Software Foundation; either
version 2.1 of the License, or (at your option) any later version.
This library is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
Library General Public License for more details.
You should have received a copy of the GNU Lesser General Public
License along with this library; if not, write to the Free
Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
*********************************************************************/
#include "public.h"
#include "basicMath.h"
#include "dynamicalMLP.h"
using namespace std;
DynamicalMLP::DynamicalMLP()
: functionType(1), neuronCount(2), mlp(0), alpha(0), beta(0)
{
type = DYN_MLP;
}
DynamicalMLP::~DynamicalMLP()
{
DEL(mlp);
}
void DynamicalMLP::Train(std::vector< std::vector<fvec> > trajectories, ivec labels)
{
if(!trajectories.size()) return;
int count = trajectories[0].size();
if(!count) return;
dim = trajectories[0][0].size()/2;
// we forget about time and just push in everything
vector<fvec> samples;
FOR(i, trajectories.size())
{
FOR(j, trajectories[i].size())
{
samples.push_back(trajectories[i][j]);
}
}
u32 sampleCnt = samples.size();
if(!sampleCnt) return;
DEL(mlp);
CvMat *layers;
// if(neuronCount == 3) neuronCount = 2; // don't ask me why but 3 neurons mess up everything...
if(!layerCount || neuronCount < 2)
{
layers = cvCreateMat(2,1,CV_32SC1);
cvSet1D(layers, 0, cvScalar(dim));
cvSet1D(layers, 1, cvScalar(dim));
}
else
{
layers = cvCreateMat(2+layerCount,1,CV_32SC1);
cvSet1D(layers, 0, cvScalar(dim));
cvSet1D(layers, layerCount+1, cvScalar(dim));
FOR(i, layerCount) cvSet1D(layers, i+1, cvScalar(neuronCount));
}
u32 *perm = randPerm(sampleCnt);
CvMat *trainSamples = cvCreateMat(sampleCnt, dim, CV_32FC1);
CvMat *trainOutputs = cvCreateMat(sampleCnt, dim, CV_32FC1);
CvMat *sampleWeights = cvCreateMat(samples.size(), 1, CV_32FC1);
FOR(i, sampleCnt)
{
FOR(j, dim) cvSetReal2D(trainSamples, i, j, samples[perm[i]][j]);
FOR(j,dim) cvSetReal2D(trainOutputs, i, j, samples[perm[i]][dim+j]);
cvSet1D(sampleWeights, i, cvScalar(1));
}
delete [] perm;
int activationFunction = functionType == 2 ? CvANN_MLP::GAUSSIAN : functionType ? CvANN_MLP::SIGMOID_SYM : CvANN_MLP::IDENTITY;
mlp = new CvANN_MLP();
mlp->create(layers, activationFunction, alpha, beta);
CvANN_MLP_TrainParams params;
params.term_crit = cvTermCriteria(CV_TERMCRIT_ITER | CV_TERMCRIT_EPS, 1000, 0.001);
mlp->train(trainSamples, trainOutputs, sampleWeights, 0, params);
cvReleaseMat(&trainSamples);
cvReleaseMat(&trainOutputs);
cvReleaseMat(&sampleWeights);
cvReleaseMat(&layers);
}
std::vector<fvec> DynamicalMLP::Test( const fvec &sample, const int count)
{
fvec start = sample;
dim = sample.size();
std::vector<fvec> res(count);
FOR(i, count) res[i].resize(dim,0);
if(!mlp) return res;
float *_input = new float[dim];
CvMat input = cvMat(1,dim,CV_32FC1, _input);
float *_output = new float[dim];
CvMat output = cvMat(1,dim,CV_32FC1, _output);
fvec velocity; velocity.resize(dim,0);
FOR(i, count)
{
res[i] = start;
start += velocity*dT;
FOR(d, dim) _input[d] = start[d];
mlp->predict(&input, &output);
FOR(d, dim) velocity[d] = _output[d];
}
delete [] _input;
delete [] _output;
return res;
}
fvec DynamicalMLP::Test( const fvec &sample)
{
int dim = sample.size();
fvec res(2);
if(!mlp) return res;
float *_input = new float[dim];
FOR(d, dim) _input[d] = sample[d];
CvMat input = cvMat(1,dim,CV_32FC1, _input);
float *_output = new float[dim];
CvMat output = cvMat(1,dim,CV_32FC1, _output);
mlp->predict(&input, &output);
FOR(d,dim) res[d] = _output[d];
delete [] _input;
delete [] _output;
return res;
}
void DynamicalMLP::SetParams(u32 functionType, u32 neuronCount, u32 layerCount, f32 alpha, f32 beta)
{
this->functionType = functionType;
this->neuronCount = neuronCount;
this->layerCount = layerCount;
this->alpha = alpha;
this->beta = beta;
}
const char *DynamicalMLP::GetInfoString()
{
char *text = new char[1024];
sprintf(text, "Multi-Layer Perceptron\n");
sprintf(text, "%sLayers: %d\n", text, layerCount);
sprintf(text, "%sNeurons: %d\n", text, neuronCount);
sprintf(text, "%sActivation Function: ", text);
switch(functionType)
{
case 0:
sprintf(text, "%s identity\n", text);
break;
case 1:
sprintf(text, "%s sigmoid (alpha: %f beta: %f)\n\t%s\n", text, alpha, beta, "beta*(1-exp(-alpha*x)) / (1 + exp(-alpha*x))");
break;
case 2:
sprintf(text, "%s gaussian (alpha: %f beta: %f)\n\t%s\n", text, alpha, beta, "beta*exp(-alpha*x*x)");
break;
}
return text;
}
|
