|
1
|
|
|
"""! |
|
2
|
|
|
|
|
3
|
|
|
@brief Cluster analysis algorithm: MBSAS (Modified Basic Sequential Algorithmic Scheme). |
|
4
|
|
|
@details Implementation based on book: |
|
5
|
|
|
- Theodoridis, Koutroumbas, Konstantinos. Elsevier Academic Press - Pattern Recognition - 2nd Edition. 2003. |
|
6
|
|
|
|
|
7
|
|
|
@authors Andrei Novikov ([email protected]) |
|
8
|
|
|
@date 2014-2018 |
|
9
|
|
|
@copyright GNU Public License |
|
10
|
|
|
|
|
11
|
|
|
@cond GNU_PUBLIC_LICENSE |
|
12
|
|
|
PyClustering is free software: you can redistribute it and/or modify |
|
13
|
|
|
it under the terms of the GNU General Public License as published by |
|
14
|
|
|
the Free Software Foundation, either version 3 of the License, or |
|
15
|
|
|
(at your option) any later version. |
|
16
|
|
|
|
|
17
|
|
|
PyClustering is distributed in the hope that it will be useful, |
|
18
|
|
|
but WITHOUT ANY WARRANTY; without even the implied warranty of |
|
19
|
|
|
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
|
20
|
|
|
GNU General Public License for more details. |
|
21
|
|
|
|
|
22
|
|
|
You should have received a copy of the GNU General Public License |
|
23
|
|
|
along with this program. If not, see <http://www.gnu.org/licenses/>. |
|
24
|
|
|
@endcond |
|
25
|
|
|
|
|
26
|
|
|
""" |
|
27
|
|
|
|
|
28
|
|
|
|
|
29
|
|
|
from pyclustering.core.mbsas_wrapper import mbsas as mbsas_wrapper; |
|
30
|
|
|
from pyclustering.core.metric_wrapper import metric_wrapper; |
|
31
|
|
|
|
|
32
|
|
|
from pyclustering.cluster.bsas import bsas; |
|
33
|
|
|
|
|
34
|
|
|
|
|
35
|
|
|
class mbsas(bsas): |
|
36
|
|
|
"""! |
|
37
|
|
|
@brief Class represents MBSAS (Modified Basic Sequential Algorithmic Scheme). |
|
38
|
|
|
@details Interface of MBSAS algorithm is the same as for BSAS. This algorithm performs clustering in two steps. |
|
39
|
|
|
The first - is determination of amount of clusters. The second - is assignment of points that were not |
|
40
|
|
|
marked as a cluster representatives to clusters. |
|
41
|
|
|
|
|
42
|
|
|
Code example of MBSAS usage: |
|
43
|
|
|
@code |
|
44
|
|
|
# Read data sample from 'Simple02.data'. |
|
45
|
|
|
sample = read_sample(SIMPLE_SAMPLES.SAMPLE_SIMPLE2); |
|
46
|
|
|
|
|
47
|
|
|
# Prepare algorithm's parameters. |
|
48
|
|
|
max_clusters = 2; |
|
49
|
|
|
threshold = 1.0; |
|
50
|
|
|
|
|
51
|
|
|
# Create instance of MBSAS algorithm. |
|
52
|
|
|
mbsas_instance = mbsas(sample, max_clusters, threshold); |
|
53
|
|
|
mbsas_instance.process(); |
|
54
|
|
|
|
|
55
|
|
|
# Get clustering results. |
|
56
|
|
|
clusters = mbsas_instance.get_clusters(); |
|
57
|
|
|
representatives = mbsas_instance.get_representatives(); |
|
58
|
|
|
|
|
59
|
|
|
# Display results. |
|
60
|
|
|
bsas_visualizer.show_clusters(sample, clusters, representatives); |
|
61
|
|
|
@endcode |
|
62
|
|
|
|
|
63
|
|
|
@see pyclustering.cluster.bsas, pyclustering.cluster.ttsas |
|
64
|
|
|
|
|
65
|
|
|
""" |
|
66
|
|
|
|
|
67
|
|
|
def __init__(self, data, maximum_clusters, threshold, ccore=True, **kwargs): |
|
68
|
|
|
"""! |
|
69
|
|
|
@brief Creates MBSAS algorithm. |
|
70
|
|
|
|
|
71
|
|
|
@param[in] data (list): Input data that is presented as list of points (objects), each point should be represented by list or tuple. |
|
72
|
|
|
@param[in] maximum_clusters: Maximum allowable number of clusters that can be allocated during processing. |
|
73
|
|
|
@param[in] threshold: Threshold of dissimilarity (maximum distance) between points. |
|
74
|
|
|
@param[in] ccore (bool): If True than DLL CCORE (C++ solution) will be used for solving. |
|
75
|
|
|
@param[in] **kwargs: Arbitrary keyword arguments (available arguments: 'metric'). |
|
76
|
|
|
|
|
77
|
|
|
Keyword Args: |
|
78
|
|
|
metric (distance_metric): Metric that is used for distance calculation between two points. |
|
79
|
|
|
|
|
80
|
|
|
""" |
|
81
|
|
|
super().__init__(data, maximum_clusters, threshold, ccore, **kwargs); |
|
82
|
|
|
|
|
83
|
|
|
|
|
84
|
|
|
def process(self): |
|
85
|
|
|
"""! |
|
86
|
|
|
@brief Performs cluster analysis in line with rules of BSAS algorithm. |
|
87
|
|
|
|
|
88
|
|
|
@remark Results of clustering can be obtained using corresponding get methods. |
|
89
|
|
|
|
|
90
|
|
|
@see get_clusters() |
|
91
|
|
|
@see get_representatives() |
|
92
|
|
|
|
|
93
|
|
|
""" |
|
94
|
|
|
|
|
95
|
|
|
if self._ccore is True: |
|
96
|
|
|
self.__process_by_ccore(); |
|
97
|
|
|
else: |
|
98
|
|
|
self.__prcess_by_python(); |
|
99
|
|
|
|
|
100
|
|
|
|
|
101
|
|
|
def __process_by_ccore(self): |
|
102
|
|
|
ccore_metric = metric_wrapper.create_instance(self._metric); |
|
103
|
|
|
self._clusters, self._representatives = mbsas_wrapper(self._data, self._amount, self._threshold, ccore_metric.get_pointer()); |
|
104
|
|
|
|
|
105
|
|
|
|
|
106
|
|
|
def __prcess_by_python(self): |
|
107
|
|
|
self._clusters.append([0]); |
|
108
|
|
|
self._representatives.append(self._data[0]); |
|
109
|
|
|
|
|
110
|
|
|
skipped_objects = []; |
|
111
|
|
|
|
|
112
|
|
|
for i in range(1, len(self._data)): |
|
113
|
|
|
point = self._data[i]; |
|
114
|
|
|
index_cluster, distance = self._find_nearest_cluster(point); |
|
115
|
|
|
|
|
116
|
|
|
if (distance > self._threshold) and (len(self._clusters) < self._amount): |
|
117
|
|
|
self._representatives.append(point); |
|
118
|
|
|
self._clusters.append([i]); |
|
119
|
|
|
else: |
|
120
|
|
|
skipped_objects.append(i); |
|
121
|
|
|
|
|
122
|
|
|
for i in skipped_objects: |
|
123
|
|
|
point = self._data[i]; |
|
124
|
|
|
index_cluster, _ = self._find_nearest_cluster(point); |
|
125
|
|
|
|
|
126
|
|
|
self._clusters[index_cluster].append(i); |
|
127
|
|
|
self._update_representative(index_cluster, point); |
annoviko /
pyclustering
