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
|
package de.lmu.ifi.dbs.elki.datasource.filter;
/*
This file is part of ELKI:
Environment for Developing KDD-Applications Supported by Index-Structures
Copyright (C) 2013
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 de.lmu.ifi.dbs.elki.data.DoubleVector;
import de.lmu.ifi.dbs.elki.data.NumberVector;
import de.lmu.ifi.dbs.elki.data.type.SimpleTypeInformation;
import de.lmu.ifi.dbs.elki.data.type.TypeUtil;
import de.lmu.ifi.dbs.elki.math.statistics.distribution.ExponentialDistribution;
import de.lmu.ifi.dbs.elki.utilities.RandomFactory;
import de.lmu.ifi.dbs.elki.utilities.documentation.Description;
import de.lmu.ifi.dbs.elki.utilities.optionhandling.AbstractParameterizer;
import de.lmu.ifi.dbs.elki.utilities.optionhandling.OptionID;
import de.lmu.ifi.dbs.elki.utilities.optionhandling.constraints.GreaterEqualConstraint;
import de.lmu.ifi.dbs.elki.utilities.optionhandling.parameterization.Parameterization;
import de.lmu.ifi.dbs.elki.utilities.optionhandling.parameters.DoubleParameter;
import de.lmu.ifi.dbs.elki.utilities.optionhandling.parameters.RandomParameter;
/**
* Add Jitter, preserving the histogram properties (same sum, nonnegative).
*
* For each vector, the total sum of all dimensions is computed.<br />
* Then a random vector of the average length <code>jitter * scale</code> is
* added and the result normalized to the original vectors sum. The individual
* dimensions are drawn from an exponential distribution with scale
* <code>jitter / dimensionality</code>, so it is expected that the error in
* most dimensions will be low, and higher in few.
*
* This is designed to degrade the quality of a histogram, while preserving the
* total sum (e.g. to keep the normalization). The factor "jitter" can be used
* to control the degradation amount.
*
* @author Erich Schubert
*
* @param <V> Vector type
*/
@Description("Add uniform Jitter to a dataset, while preserving the total vector sum.")
public class HistogramJitterFilter<V extends NumberVector<?>> extends AbstractVectorStreamConversionFilter<V, V> {
/**
* Jitter amount.
*/
double jitter;
/**
* Random generator.
*/
ExponentialDistribution rnd;
/**
* Constructor.
*
* @param jitter Relative amount of jitter to add
* @param rnd Random generator
*/
public HistogramJitterFilter(double jitter, RandomFactory rnd) {
super();
this.jitter = jitter;
this.rnd = new ExponentialDistribution(1, rnd.getRandom());
}
@Override
protected V filterSingleObject(V obj) {
final int dim = obj.getDimensionality();
// Compute the total sum.
double osum = 0;
for (int i = 0; i < dim; i++) {
osum += obj.doubleValue(i);
}
// Actual maximum jitter amount:
final double maxjitter = 2 * jitter / dim * osum;
// Generate jitter vector
double[] raw = new double[dim];
double jsum = 0; // Sum of jitter
for (int i = 0; i < raw.length; i++) {
raw[i] = rnd.nextRandom() * maxjitter;
jsum += raw[i];
}
final double mix = jsum / osum;
// Combine the two vector
for (int i = 0; i < raw.length; i++) {
raw[i] = raw[i] + (1 - mix) * obj.doubleValue(i);
}
return factory.newNumberVector(raw);
}
@Override
protected SimpleTypeInformation<? super V> getInputTypeRestriction() {
return TypeUtil.NUMBER_VECTOR_VARIABLE_LENGTH;
}
@Override
protected SimpleTypeInformation<V> convertedType(SimpleTypeInformation<V> in) {
initializeOutputType(in);
return in;
}
/**
* Parameterization class.
*
* @author Erich Schubert
*
* @apiviz.exclude
*/
public static class Parameterizer extends AbstractParameterizer {
/**
* Option ID for the jitter strength.
*/
public static final OptionID JITTER_ID = new OptionID("jitter.amount", "Jitter amount relative to data.");
/**
* Option ID for the jitter random seed.
*/
public static final OptionID SEED_ID = new OptionID("jitter.seed", "Jitter random seed.");
/**
* Jitter amount.
*/
double jitter = 0.1;
/**
* Random generator seed.
*/
RandomFactory rnd;
@Override
protected void makeOptions(Parameterization config) {
super.makeOptions(config);
DoubleParameter jitterP = new DoubleParameter(JITTER_ID);
jitterP.addConstraint(new GreaterEqualConstraint(Double.valueOf(0.0)));
if (config.grab(jitterP)) {
jitter = jitterP.getValue().doubleValue();
}
RandomParameter rndP = new RandomParameter(SEED_ID);
if (config.grab(rndP)) {
rnd = rndP.getValue();
}
}
@Override
protected HistogramJitterFilter<DoubleVector> makeInstance() {
return new HistogramJitterFilter<>(jitter, rnd);
}
}
}
|
