Cluster.java
3.78 KB
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
/**
*
*/
package is2.data;
import java.io.BufferedReader;
import java.io.DataInputStream;
import java.io.DataOutputStream;
import java.io.FileInputStream;
import java.io.IOException;
import java.io.InputStreamReader;
import is2.util.DB;
/**
* @author Dr. Bernd Bohnet, 28.10.2010
*
*
*/
final public class Cluster {
public static final String LPATH = "LP";
public static final String SPATH = "SP";
// [word][p] p = [0:long-path | 1:short-path]
final private short[][] word2path;
public Cluster() {
word2path = new short[0][0];
}
/**
* @param clusterFile
* @param mf
*
*/
public Cluster(String clusterFile, IEncoderPlus mf, int ls) {
final String REGEX = "\t";
// register words
try {
BufferedReader inputReader = new BufferedReader(
new InputStreamReader(new FileInputStream(clusterFile), "UTF-8"), 32768);
int cnt = 0;
String line;
while ((line = inputReader.readLine()) != null) {
cnt++;
try {
String[] split = line.split(REGEX);
mf.register(SPATH, split[0].length() < ls ? split[0] : split[0].substring(0, ls));
mf.register(LPATH, split[0]);
mf.register(PipeGen.WORD, split[1]);
} catch (Exception e) {
System.out.println("Error in cluster line " + cnt + " error: " + e.getMessage());
}
}
System.out.println("read number of clusters " + cnt);
inputReader.close();
} catch (Exception e) {
e.printStackTrace();
}
word2path = new short[mf.getFeatureCounter().get(PipeGen.WORD)][2];
// insert words
try {
String line;
BufferedReader inputReader = new BufferedReader(
new InputStreamReader(new FileInputStream(clusterFile), "UTF-8"), 32768);
while ((line = inputReader.readLine()) != null) {
String[] split = line.split(REGEX);
int wd = mf.getValue(PipeGen.WORD, split[1]);
word2path[wd][0] = (short) mf.getValue(SPATH,
split[0].length() < ls ? split[0] : split[0].substring(0, ls));
word2path[wd][1] = (short) mf.getValue(LPATH, split[0]);
}
inputReader.close();
int fill = 0;
for (short[] element : word2path) {
if (element[0] != 0)
fill++;
}
/*
* for(int l = 0; l<word2path.length; l++ ){ if (word2path[l][1]!=0)
* fillL++; if (word2path[l][1]<-1)
* System.out.println("lower "+word2path[l][1]); }
*/
System.out.println("filled " + fill + " of " + word2path.length);
} catch (Exception e) {
e.printStackTrace();
}
}
/**
* Read the cluster
*
* @param dos
* @throws IOException
*/
public Cluster(DataInputStream dis) throws IOException {
word2path = new short[dis.readInt()][2];
for (int i = 0; i < word2path.length; i++) {
word2path[i][0] = dis.readShort();
word2path[i][1] = dis.readShort();
}
DB.println("Read cluster with " + word2path.length + " words ");
}
/**
* Write the cluster
*
* @param dos
* @throws IOException
*/
public void write(DataOutputStream dos) throws IOException {
dos.writeInt(word2path.length);
for (short[] i : word2path) {
dos.writeShort(i[0]);
dos.writeShort(i[1]);
}
}
/**
* @param form
* the id of a word form
* @return the short path to the word form in the cluster
*
* final public int getSP(int form) { if (word2path.length<form)
* return -1; return word2path[form][0]; }
*/
/**
* get the long path to a word form in the cluster
*
* @param form
* the id of a word form
* @return the long path to the word
*/
final public int getLP(int form) {
if (word2path.length <= form || word2path[form].length <= 0)
return -1;
return word2path[form][0] == 0 ? -1 : word2path[form][0];
}
final public int getLP(int form, int l) {
if (word2path.length < form)
return -1;
return word2path[form][l] == 0 ? -1 : word2path[form][l];
}
final public int size() {
return word2path.length;
}
}