|
1
|
|
|
"""!
|
|
2
|
|
|
|
|
3
|
|
|
@brief Cluster analysis algorithm: BANG.
|
|
4
|
|
|
@details Implementation based on paper @cite inproceedings::bang::1.
|
|
5
|
|
|
|
|
6
|
|
|
@authors Andrei Novikov ([email protected])
|
|
7
|
|
|
@date 2014-2018
|
|
8
|
|
|
@copyright GNU Public License
|
|
9
|
|
|
|
|
10
|
|
|
@cond GNU_PUBLIC_LICENSE
|
|
11
|
|
|
PyClustering is free software: you can redistribute it and/or modify
|
|
12
|
|
|
it under the terms of the GNU General Public License as published by
|
|
13
|
|
|
the Free Software Foundation, either version 3 of the License, or
|
|
14
|
|
|
(at your option) any later version.
|
|
15
|
|
|
|
|
16
|
|
|
PyClustering is distributed in the hope that it will be useful,
|
|
17
|
|
|
but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
18
|
|
|
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
19
|
|
|
GNU General Public License for more details.
|
|
20
|
|
|
|
|
21
|
|
|
You should have received a copy of the GNU General Public License
|
|
22
|
|
|
along with this program. If not, see <http://www.gnu.org/licenses/>.
|
|
23
|
|
|
@endcond
|
|
24
|
|
|
|
|
25
|
|
|
"""
|
|
26
|
|
|
|
|
27
|
|
|
|
|
28
|
|
|
from pyclustering.utils import data_corners
|
|
29
|
|
|
|
|
30
|
|
|
|
|
31
|
|
|
class bang_block:
|
|
32
|
|
|
def __init__(self, data, dimension, region, level, max_corner, min_corner):
|
|
33
|
|
|
self.__region_number = region
|
|
34
|
|
|
self.__level = level
|
|
35
|
|
|
self.__data = data
|
|
36
|
|
|
self.__dimension = dimension
|
|
37
|
|
|
self.__max_corner = max_corner
|
|
38
|
|
|
self.__min_corner = min_corner
|
|
39
|
|
|
self.__density = self.__calculate_density()
|
|
40
|
|
|
|
|
41
|
|
|
|
|
42
|
|
|
def get_region(self):
|
|
43
|
|
|
return self.__region_number
|
|
44
|
|
|
|
|
45
|
|
|
|
|
46
|
|
|
def get_level(self):
|
|
47
|
|
|
return self.__level
|
|
48
|
|
|
|
|
49
|
|
|
|
|
50
|
|
|
def get_density(self):
|
|
51
|
|
|
return self.__density
|
|
52
|
|
|
|
|
53
|
|
|
|
|
54
|
|
|
def split(self):
|
|
55
|
|
|
left_region_number = self.__region_number
|
|
56
|
|
|
right_region_number = self.__region_number + 2 ** self.__level
|
|
57
|
|
|
|
|
58
|
|
|
dimension = self.__dimension + 1
|
|
59
|
|
|
if dimension > len(self.__data[0]):
|
|
60
|
|
|
dimension = 0
|
|
61
|
|
|
|
|
62
|
|
|
left_min_corner = self.__min_corner[:]
|
|
63
|
|
|
right_max_corner = self.__max_corner[:]
|
|
64
|
|
|
|
|
65
|
|
|
left_min_corner[dimension] += (self.__max_corner[dimension] - self.__min_corner[dimension]) / 2.0
|
|
66
|
|
|
right_max_corner[dimension] = left_min_corner[dimension]
|
|
67
|
|
|
|
|
68
|
|
|
left = bang_block(self.__data, dimension, left_region_number, self.__level + 1, self.__max_corner, left_min_corner)
|
|
69
|
|
|
right = bang_block(self.__data, dimension, right_region_number, self.__level + 1, right_max_corner, self.__min_corner)
|
|
70
|
|
|
|
|
71
|
|
|
return left, right
|
|
72
|
|
|
|
|
73
|
|
|
|
|
74
|
|
|
def contained(self, point):
|
|
75
|
|
|
for i in range(len(point)):
|
|
76
|
|
|
if point[i] <= self.__min_corner[i] or point[i] > self.__max_corner[i]:
|
|
77
|
|
|
return False
|
|
78
|
|
|
|
|
79
|
|
|
|
|
80
|
|
|
def __calculate_density(self):
|
|
81
|
|
|
volume = self.__max_corner[0] - self.__min_corner[0]
|
|
82
|
|
|
for i in range(1, len(self.__max_corner)):
|
|
83
|
|
|
volume *= self.__max_corner[i] - self.__min_corner[i]
|
|
84
|
|
|
|
|
85
|
|
|
amount = self.__get_amount_points()
|
|
86
|
|
|
return amount / volume
|
|
87
|
|
|
|
|
88
|
|
|
|
|
89
|
|
|
def __get_amount_points(self):
|
|
90
|
|
|
amount = 0
|
|
91
|
|
|
for point in self.__data:
|
|
92
|
|
|
if self.contained(point):
|
|
93
|
|
|
amount += 1
|
|
94
|
|
|
|
|
95
|
|
|
return amount
|
|
96
|
|
|
|
|
97
|
|
|
|
|
98
|
|
|
|
|
99
|
|
|
class bang:
|
|
100
|
|
|
def __init__(self, data, levels, density_threshold = 0.0):
|
|
101
|
|
|
self.__data = data
|
|
102
|
|
|
self.__levels = levels
|
|
103
|
|
|
self.__blocks = []
|
|
104
|
|
|
self.__density_threshold = density_threshold
|
|
105
|
|
|
|
|
106
|
|
|
|
|
107
|
|
|
def process(self):
|
|
108
|
|
|
self.__build_blocks()
|
|
109
|
|
|
self.__blocks = sorted(self.__blocks, key=lambda block: block.get_density())
|
|
110
|
|
|
|
|
111
|
|
|
|
|
112
|
|
|
def get_blocks(self):
|
|
113
|
|
|
return self.__blocks
|
|
114
|
|
|
|
|
115
|
|
|
|
|
116
|
|
|
def __build_blocks(self):
|
|
117
|
|
|
max_corner, min_corner = data_corners(self.__data)
|
|
118
|
|
|
root_block = bang_block(self.__data, 0, 0, 0, max_corner, min_corner)
|
|
119
|
|
|
|
|
120
|
|
|
self.__blocks.append(root_block)
|
|
121
|
|
|
level_blocks = [root_block]
|
|
122
|
|
|
|
|
123
|
|
|
for level in range(self.__levels):
|
|
|
|
|
|
|
124
|
|
|
level_blocks = self.__build_next_level_blocks(level_blocks)
|
|
125
|
|
|
|
|
126
|
|
|
|
|
127
|
|
|
def __build_next_level_blocks(self, level_blocks):
|
|
128
|
|
|
next_level_blocks = []
|
|
129
|
|
|
for block in level_blocks:
|
|
130
|
|
|
left, right = block.split()
|
|
131
|
|
|
|
|
132
|
|
|
next_level_blocks.append(left)
|
|
133
|
|
|
next_level_blocks.append(right)
|
|
134
|
|
|
|
|
135
|
|
|
return next_level_blocks |
annoviko /
pyclustering
Loading history...