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
|
package de.lmu.ifi.dbs.elki.math.linearalgebra.randomprojections;
/*
This file is part of ELKI:
Environment for Developing KDD-Applications Supported by Index-Structures
Copyright (C) 2015
Ludwig-Maximilians-Universität München
Lehr- und Forschungseinheit für Datenbanksysteme
ELKI Development Team
This program is free software: you can redistribute it and/or modify
it under the terms of the GNU Affero General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
This program 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 Affero General Public License for more details.
You should have received a copy of the GNU Affero General Public License
along with this program. If not, see <http://www.gnu.org/licenses/>.
*/
import java.util.Arrays;
import java.util.Random;
import de.lmu.ifi.dbs.elki.data.NumberVector;
import de.lmu.ifi.dbs.elki.math.MathUtil;
import de.lmu.ifi.dbs.elki.math.random.RandomFactory;
import de.lmu.ifi.dbs.elki.utilities.documentation.Reference;
/**
* Random projection family based on selecting random features.
*
* The basic idea of using this for data mining should probably be attributed to
* L. Breiman, who used it to improve the performance of predictors in an
* ensemble.
*
* Reference:
* <p>
* L. Breiman<br />
* Bagging predictors<br />
* Machine learning 24.2
* </p>
*
* @author Erich Schubert
* @since 0.6.0
*/
@Reference(authors = "L. Breiman", //
title = "Bagging predictors", //
booktitle = "Machine learning 24.2", //
url = "http://dx.doi.org/10.1007/BF00058655")
public class RandomSubsetProjectionFamily extends AbstractRandomProjectionFamily {
/**
* Constructor.
*
* @param random Random generator.
*/
public RandomSubsetProjectionFamily(RandomFactory random) {
super(random);
}
@Override
public Projection generateProjection(int idim, int odim) {
int[] dims;
if(odim < idim) {
dims = Arrays.copyOf(randomPermutation(MathUtil.sequence(0, idim), random), odim);
}
else if(odim == idim) {
dims = randomPermutation(MathUtil.sequence(0, idim), random);
}
else {
int mdim = idim;
while(mdim < odim) {
mdim += idim;
}
dims = new int[mdim];
for(int i = 0; i < mdim; i++) {
dims[i] = i % idim;
}
dims = Arrays.copyOf(randomPermutation(dims, random), odim);
}
return new SubsetProjection(dims);
}
/**
* Perform a random permutation of the array, in-place.
*
* Knuth / Fisher-Yates style shuffle
*
* FIXME: move to shared code.
*
* @param out Existing output array
* @param random Random generator.
* @return Same array.
*/
public static int[] randomPermutation(final int[] out, Random random) {
for(int i = out.length - 1; i > 0; i--) {
// Swap with random preceeding element.
int ri = random.nextInt(i + 1);
int tmp = out[ri];
out[ri] = out[i];
out[i] = tmp;
}
return out;
}
/**
* Random subset projection.
*
* @author Erich Schubert
*/
public static class SubsetProjection implements Projection {
/**
* Input dimensions.
*/
private int[] dims;
/**
* Constructor.
*
* @param dims Data permutation.
*/
public SubsetProjection(int[] dims) {
this.dims = dims;
}
@Override
public double[] project(NumberVector in) {
return project(in, new double[dims.length]);
}
@Override
public double[] project(NumberVector in, double[] buffer) {
for(int i = 0; i < dims.length; i++) {
buffer[i] = in.doubleValue(dims[i]);
}
return buffer;
}
@Override
public int getOutputDimensionality() {
return dims.length;
}
}
/**
* Parameterization class.
*
* @author Erich Schubert
*
* @apiviz.exclude
*/
public static class Parameterizer extends AbstractRandomProjectionFamily.Parameterizer {
@Override
protected RandomSubsetProjectionFamily makeInstance() {
return new RandomSubsetProjectionFamily(random);
}
}
}
|
