|
1
|
|
|
# -*- coding: utf-8 -*- |
|
2
|
|
|
|
|
3
|
|
|
# Copyright 2014-2018 by Christopher C. Little. |
|
4
|
|
|
# This file is part of Abydos. |
|
5
|
|
|
# |
|
6
|
|
|
# Abydos is free software: you can redistribute it and/or modify |
|
7
|
|
|
# it under the terms of the GNU General Public License as published by |
|
8
|
|
|
# the Free Software Foundation, either version 3 of the License, or |
|
9
|
|
|
# (at your option) any later version. |
|
10
|
|
|
# |
|
11
|
|
|
# Abydos is distributed in the hope that it will be useful, |
|
12
|
|
|
# but WITHOUT ANY WARRANTY; without even the implied warranty of |
|
13
|
|
|
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
|
14
|
|
|
# GNU General Public License for more details. |
|
15
|
|
|
# |
|
16
|
|
|
# You should have received a copy of the GNU General Public License |
|
17
|
|
|
# along with Abydos. If not, see <http://www.gnu.org/licenses/>. |
|
18
|
|
|
|
|
19
|
1 |
|
"""abydos.distance._bag. |
|
20
|
|
|
|
|
21
|
|
|
Bag similarity & distance |
|
22
|
|
|
""" |
|
23
|
|
|
|
|
24
|
1 |
|
from __future__ import ( |
|
25
|
|
|
absolute_import, |
|
26
|
|
|
division, |
|
27
|
|
|
print_function, |
|
28
|
|
|
unicode_literals, |
|
29
|
|
|
) |
|
30
|
|
|
|
|
31
|
1 |
|
from collections import Counter |
|
32
|
|
|
|
|
33
|
1 |
|
from ._token_distance import _TokenDistance |
|
34
|
|
|
|
|
35
|
1 |
|
__all__ = ['Bag', 'bag', 'dist_bag', 'sim_bag'] |
|
36
|
|
|
|
|
37
|
|
|
|
|
38
|
1 |
|
class Bag(_TokenDistance): |
|
|
|
|
|
|
39
|
|
|
"""Bag distance. |
|
40
|
|
|
|
|
41
|
|
|
Bag distance is proposed in :cite:`Bartolini:2002`. It is defined as: |
|
42
|
|
|
:math:`max(|multiset(src)-multiset(tar)|, |multiset(tar)-multiset(src)|)`. |
|
43
|
|
|
""" |
|
44
|
|
|
|
|
45
|
1 |
|
def dist_abs(self, src, tar): |
|
|
|
|
|
|
46
|
|
|
"""Return the bag distance between two strings. |
|
47
|
|
|
|
|
48
|
|
|
Parameters |
|
49
|
|
|
---------- |
|
50
|
|
|
src : str |
|
51
|
|
|
Source string for comparison |
|
52
|
|
|
tar : str |
|
53
|
|
|
Target string for comparison |
|
54
|
|
|
|
|
55
|
|
|
Returns |
|
56
|
|
|
------- |
|
57
|
|
|
int |
|
58
|
|
|
Bag distance |
|
59
|
|
|
|
|
60
|
|
|
Examples |
|
61
|
|
|
-------- |
|
62
|
|
|
>>> cmp = Bag() |
|
63
|
|
|
>>> cmp.dist_abs('cat', 'hat') |
|
64
|
|
|
1 |
|
65
|
|
|
>>> cmp.dist_abs('Niall', 'Neil') |
|
66
|
|
|
2 |
|
67
|
|
|
>>> cmp.dist_abs('aluminum', 'Catalan') |
|
68
|
|
|
5 |
|
69
|
|
|
>>> cmp.dist_abs('ATCG', 'TAGC') |
|
70
|
|
|
0 |
|
71
|
|
|
>>> cmp.dist_abs('abcdefg', 'hijklm') |
|
72
|
|
|
7 |
|
73
|
|
|
>>> cmp.dist_abs('abcdefg', 'hijklmno') |
|
74
|
|
|
8 |
|
75
|
|
|
|
|
76
|
|
|
""" |
|
77
|
1 |
|
if tar == src: |
|
78
|
1 |
|
return 0 |
|
79
|
1 |
|
elif not src: |
|
80
|
1 |
|
return len(tar) |
|
81
|
1 |
|
elif not tar: |
|
82
|
1 |
|
return len(src) |
|
83
|
|
|
|
|
84
|
1 |
|
src_bag = Counter(src) |
|
85
|
1 |
|
tar_bag = Counter(tar) |
|
86
|
1 |
|
return max( |
|
87
|
|
|
sum((src_bag - tar_bag).values()), |
|
88
|
|
|
sum((tar_bag - src_bag).values()), |
|
89
|
|
|
) |
|
90
|
|
|
|
|
91
|
1 |
|
def dist(self, src, tar): |
|
|
|
|
|
|
92
|
|
|
"""Return the normalized bag distance between two strings. |
|
93
|
|
|
|
|
94
|
|
|
Bag distance is normalized by dividing by :math:`max( |src|, |tar| )`. |
|
95
|
|
|
|
|
96
|
|
|
Parameters |
|
97
|
|
|
---------- |
|
98
|
|
|
src : str |
|
99
|
|
|
Source string for comparison |
|
100
|
|
|
tar : str |
|
101
|
|
|
Target string for comparison |
|
102
|
|
|
|
|
103
|
|
|
Returns |
|
104
|
|
|
------- |
|
105
|
|
|
float |
|
106
|
|
|
Normalized bag distance |
|
107
|
|
|
|
|
108
|
|
|
Examples |
|
109
|
|
|
-------- |
|
110
|
|
|
>>> cmp = Bag() |
|
111
|
|
|
>>> cmp.dist('cat', 'hat') |
|
112
|
|
|
0.3333333333333333 |
|
113
|
|
|
>>> cmp.dist('Niall', 'Neil') |
|
114
|
|
|
0.4 |
|
115
|
|
|
>>> cmp.dist('aluminum', 'Catalan') |
|
116
|
|
|
0.625 |
|
117
|
|
|
>>> cmp.dist('ATCG', 'TAGC') |
|
118
|
|
|
0.0 |
|
119
|
|
|
|
|
120
|
|
|
""" |
|
121
|
1 |
|
if tar == src: |
|
122
|
1 |
|
return 0.0 |
|
123
|
1 |
|
if not src or not tar: |
|
124
|
1 |
|
return 1.0 |
|
125
|
|
|
|
|
126
|
1 |
|
max_length = max(len(src), len(tar)) |
|
127
|
|
|
|
|
128
|
1 |
|
return self.dist_abs(src, tar) / max_length |
|
129
|
|
|
|
|
130
|
|
|
|
|
131
|
1 |
|
def bag(src, tar): |
|
132
|
|
|
"""Return the bag distance between two strings. |
|
133
|
|
|
|
|
134
|
|
|
This is a wrapper for :py:meth:`Bag.dist_abs`. |
|
135
|
|
|
|
|
136
|
|
|
Parameters |
|
137
|
|
|
---------- |
|
138
|
|
|
src : str |
|
139
|
|
|
Source string for comparison |
|
140
|
|
|
tar : str |
|
141
|
|
|
Target string for comparison |
|
142
|
|
|
|
|
143
|
|
|
Returns |
|
144
|
|
|
------- |
|
145
|
|
|
int |
|
146
|
|
|
Bag distance |
|
147
|
|
|
|
|
148
|
|
|
Examples |
|
149
|
|
|
-------- |
|
150
|
|
|
>>> bag('cat', 'hat') |
|
151
|
|
|
1 |
|
152
|
|
|
>>> bag('Niall', 'Neil') |
|
153
|
|
|
2 |
|
154
|
|
|
>>> bag('aluminum', 'Catalan') |
|
155
|
|
|
5 |
|
156
|
|
|
>>> bag('ATCG', 'TAGC') |
|
157
|
|
|
0 |
|
158
|
|
|
>>> bag('abcdefg', 'hijklm') |
|
159
|
|
|
7 |
|
160
|
|
|
>>> bag('abcdefg', 'hijklmno') |
|
161
|
|
|
8 |
|
162
|
|
|
|
|
163
|
|
|
""" |
|
164
|
1 |
|
return Bag().dist_abs(src, tar) |
|
165
|
|
|
|
|
166
|
|
|
|
|
167
|
1 |
|
def dist_bag(src, tar): |
|
168
|
|
|
"""Return the normalized bag distance between two strings. |
|
169
|
|
|
|
|
170
|
|
|
This is a wrapper for :py:meth:`Bag.dist`. |
|
171
|
|
|
|
|
172
|
|
|
Parameters |
|
173
|
|
|
---------- |
|
174
|
|
|
src : str |
|
175
|
|
|
Source string for comparison |
|
176
|
|
|
tar : str |
|
177
|
|
|
Target string for comparison |
|
178
|
|
|
|
|
179
|
|
|
Returns |
|
180
|
|
|
------- |
|
181
|
|
|
float |
|
182
|
|
|
Normalized bag distance |
|
183
|
|
|
|
|
184
|
|
|
Examples |
|
185
|
|
|
-------- |
|
186
|
|
|
>>> dist_bag('cat', 'hat') |
|
187
|
|
|
0.3333333333333333 |
|
188
|
|
|
>>> dist_bag('Niall', 'Neil') |
|
189
|
|
|
0.4 |
|
190
|
|
|
>>> dist_bag('aluminum', 'Catalan') |
|
191
|
|
|
0.625 |
|
192
|
|
|
>>> dist_bag('ATCG', 'TAGC') |
|
193
|
|
|
0.0 |
|
194
|
|
|
|
|
195
|
|
|
""" |
|
196
|
1 |
|
return Bag().dist(src, tar) |
|
197
|
|
|
|
|
198
|
|
|
|
|
199
|
1 |
|
def sim_bag(src, tar): |
|
200
|
|
|
"""Return the normalized bag similarity of two strings. |
|
201
|
|
|
|
|
202
|
|
|
This is a wrapper for :py:meth:`Bag.sim`. |
|
203
|
|
|
|
|
204
|
|
|
Parameters |
|
205
|
|
|
---------- |
|
206
|
|
|
src : str |
|
207
|
|
|
Source string for comparison |
|
208
|
|
|
tar : str |
|
209
|
|
|
Target string for comparison |
|
210
|
|
|
|
|
211
|
|
|
Returns |
|
212
|
|
|
------- |
|
213
|
|
|
float |
|
214
|
|
|
Normalized bag similarity |
|
215
|
|
|
|
|
216
|
|
|
Examples |
|
217
|
|
|
-------- |
|
218
|
|
|
>>> round(sim_bag('cat', 'hat'), 12) |
|
219
|
|
|
0.666666666667 |
|
220
|
|
|
>>> sim_bag('Niall', 'Neil') |
|
221
|
|
|
0.6 |
|
222
|
|
|
>>> sim_bag('aluminum', 'Catalan') |
|
223
|
|
|
0.375 |
|
224
|
|
|
>>> sim_bag('ATCG', 'TAGC') |
|
225
|
|
|
1.0 |
|
226
|
|
|
|
|
227
|
|
|
""" |
|
228
|
1 |
|
return Bag().sim(src, tar) |
|
229
|
|
|
|
|
230
|
|
|
|
|
231
|
|
|
if __name__ == '__main__': |
|
232
|
|
|
import doctest |
|
233
|
|
|
|
|
234
|
|
|
doctest.testmod() |
|
235
|
|
|
|
chrislit /
abydos
Loading history...