|
1
|
|
|
# encoding=utf8 |
|
2
|
|
|
# pylint: disable=mixed-indentation, multiple-statements, attribute-defined-outside-init, logging-not-lazy, no-self-use, line-too-long, singleton-comparison, arguments-differ, bad-continuation |
|
3
|
|
|
import logging |
|
4
|
|
|
from numpy import full |
|
5
|
|
|
from NiaPy.algorithms.algorithm import Algorithm |
|
6
|
|
|
|
|
7
|
|
|
logging.basicConfig() |
|
8
|
|
|
logger = logging.getLogger('NiaPy.algorithms.basic') |
|
9
|
|
|
logger.setLevel('INFO') |
|
10
|
|
|
|
|
11
|
|
|
__all__ = ['BatAlgorithm'] |
|
12
|
|
|
|
|
13
|
|
|
class BatAlgorithm(Algorithm): |
|
14
|
|
|
r"""Implementation of Bat algorithm. |
|
15
|
|
|
|
|
16
|
|
|
Algorithm: |
|
17
|
|
|
Bat algorithm |
|
18
|
|
|
|
|
19
|
|
|
Date: |
|
20
|
|
|
2015 |
|
21
|
|
|
|
|
22
|
|
|
Authors: |
|
23
|
|
|
Iztok Fister Jr., Marko Burjek and Klemen Berkovič |
|
24
|
|
|
|
|
25
|
|
|
License: |
|
26
|
|
|
MIT |
|
27
|
|
|
|
|
28
|
|
|
Reference paper: |
|
29
|
|
|
Yang, Xin-She. "A new metaheuristic bat-inspired algorithm." Nature inspired cooperative strategies for optimization (NICSO 2010). Springer, Berlin, Heidelberg, 2010. 65-74. |
|
30
|
|
|
|
|
31
|
|
|
Attributes: |
|
32
|
|
|
Name (List[str]): List of strings representing algorithm name. |
|
33
|
|
|
A (float): Loudness. |
|
34
|
|
|
r (float): Pulse rate. |
|
35
|
|
|
Qmin (float): Minimum frequency. |
|
36
|
|
|
Qmax (float): Maximum frequency. |
|
37
|
|
|
|
|
38
|
|
|
See Also: |
|
39
|
|
|
* :class:`NiaPy.algorithms.Algorithm` |
|
40
|
|
|
""" |
|
41
|
|
|
Name = ['BatAlgorithm', 'BA'] |
|
42
|
|
|
|
|
43
|
|
View Code Duplication |
@staticmethod |
|
|
|
|
|
|
44
|
|
|
def typeParameters(): |
|
45
|
|
|
r"""Return dict with where key of dict represents parameter name and values represent checking functions for selected parameter. |
|
46
|
|
|
|
|
47
|
|
|
Returns: |
|
48
|
|
|
Dict[str, Callable]: |
|
49
|
|
|
* A (Callable[[Union[float, int]], bool]): Loudness. |
|
50
|
|
|
* r (Callable[[Union[float, int]], bool]): Pulse rate. |
|
51
|
|
|
* Qmin (Callable[[Union[float, int]], bool]): Minimum frequency. |
|
52
|
|
|
* Qmax (Callable[[Union[float, int]], bool]): Maximum frequency. |
|
53
|
|
|
|
|
54
|
|
|
See Also: |
|
55
|
|
|
* :func:`NiaPy.algorithms.Algorithm.typeParameters` |
|
56
|
|
|
""" |
|
57
|
|
|
d = Algorithm.typeParameters() |
|
58
|
|
|
d.update({ |
|
59
|
|
|
'A': lambda x: isinstance(x, (float, int)) and x > 0, |
|
60
|
|
|
'r': lambda x: isinstance(x, (float, int)) and x > 0, |
|
61
|
|
|
'Qmin': lambda x: isinstance(x, (float, int)), |
|
62
|
|
|
'Qmax': lambda x: isinstance(x, (float, int)) |
|
63
|
|
|
}) |
|
64
|
|
|
return d |
|
65
|
|
|
|
|
66
|
|
|
def setParameters(self, NP=40, A=0.5, r=0.5, Qmin=0.0, Qmax=2.0, **ukwargs): |
|
67
|
|
|
r"""Set the parameters of the algorithm. |
|
68
|
|
|
|
|
69
|
|
|
Args: |
|
70
|
|
|
A (Optional[float]): Loudness. |
|
71
|
|
|
r (Optional[float]): Pulse rate. |
|
72
|
|
|
Qmin (Optional[float]): Minimum frequency. |
|
73
|
|
|
Qmax (Optional[float]): Maximum frequency. |
|
74
|
|
|
|
|
75
|
|
|
See Also: |
|
76
|
|
|
* :func:`NiaPy.algorithms.Algorithm.setParameters` |
|
77
|
|
|
""" |
|
78
|
|
|
Algorithm.setParameters(self, NP=NP, **ukwargs) |
|
79
|
|
|
self.A, self.r, self.Qmin, self.Qmax = A, r, Qmin, Qmax |
|
80
|
|
|
if ukwargs: logger.info('Unused arguments: %s' % (ukwargs)) |
|
81
|
|
|
|
|
82
|
|
|
def initPopulation(self, task): |
|
83
|
|
|
r"""Initialize the starting population. |
|
84
|
|
|
|
|
85
|
|
|
Parameters: |
|
86
|
|
|
task (Task): Optimization task |
|
87
|
|
|
|
|
88
|
|
|
Returns: |
|
89
|
|
|
Tuple[numpy.ndarray, numpy.ndarray[float], Dict[str, Any]]: |
|
90
|
|
|
1. New population. |
|
91
|
|
|
2. New population fitness/function values. |
|
92
|
|
|
3. Additional arguments: |
|
93
|
|
|
* S (numpy.ndarray): TODO |
|
94
|
|
|
* Q (numpy.ndarray[float]): TODO |
|
95
|
|
|
* v (numpy.ndarray[float]): TODO |
|
96
|
|
|
|
|
97
|
|
|
See Also: |
|
98
|
|
|
* :func:`NiaPy.algorithms.Algorithm.initPopulation` |
|
99
|
|
|
""" |
|
100
|
|
|
Sol, Fitness, d = Algorithm.initPopulation(self, task) |
|
101
|
|
|
S, Q, v = full([self.NP, task.D], 0.0), full(self.NP, 0.0), full([self.NP, task.D], 0.0) |
|
102
|
|
|
d.update({'S': S, 'Q': Q, 'v': v}) |
|
103
|
|
|
return Sol, Fitness, d |
|
104
|
|
|
|
|
105
|
|
|
def generateBest(self, best, task, **kwargs): |
|
106
|
|
|
r"""Generate new solution based on global best known solution. |
|
107
|
|
|
|
|
108
|
|
|
Args: |
|
109
|
|
|
best (numpy.ndarray): Global best individual. |
|
110
|
|
|
task (Task): Optimization task. |
|
111
|
|
|
**kwargs (Dict[str, Any]): Additional arguments. |
|
112
|
|
|
|
|
113
|
|
|
Returns: |
|
114
|
|
|
numpy.ndarray: New solution based on global best individual. |
|
115
|
|
|
""" |
|
116
|
|
|
return task.repair(best + 0.001 * self.normal(0, 1, task.D)) |
|
117
|
|
|
|
|
118
|
|
|
def runIteration(self, task, Sol, Fitness, best, f_min, S, Q, v, **dparams): |
|
119
|
|
|
r"""Core function of Bat Algorithm. |
|
120
|
|
|
|
|
121
|
|
|
Parameters: |
|
122
|
|
|
task (Task): Optimization task. |
|
123
|
|
|
Sol (numpy.ndarray): Current population |
|
124
|
|
|
Fitness (numpy.ndarray[float]): Current population fitness/funciton values |
|
125
|
|
|
best (numpy.ndarray): Current best individual |
|
126
|
|
|
f_min (float): Current best individual function/fitness value |
|
127
|
|
|
S (numpy.ndarray): TODO |
|
128
|
|
|
Q (numpy.ndarray[float]): TODO |
|
129
|
|
|
v (numpy.ndarray[float]): TODO |
|
130
|
|
|
dparams (Dict[str, Any]): Additional algorithm arguments |
|
131
|
|
|
|
|
132
|
|
|
Returns: |
|
133
|
|
|
Tuple[numpy.ndarray, numpy.ndarray[float], Dict[str, Any]]: |
|
134
|
|
|
1. New population |
|
135
|
|
|
2. New population fitness/function vlues |
|
136
|
|
|
3. Additional arguments: |
|
137
|
|
|
* S (numpy.ndarray): TODO |
|
138
|
|
|
* Q (numpy.ndarray[float]): TODO |
|
139
|
|
|
* v (numpy.ndarray[float]): TODO |
|
140
|
|
|
""" |
|
141
|
|
|
for i in range(self.NP): |
|
142
|
|
|
Q[i] = self.Qmin + (self.Qmax - self.Qmin) * self.uniform(0, 1) |
|
143
|
|
|
v[i] += (Sol[i] - best) * Q[i] |
|
144
|
|
|
S[i] = task.repair(Sol[i] + v[i]) |
|
145
|
|
|
if self.rand() > self.r: S[i] = self.generateBest(best=best, task=task, i=i, Sol=Sol) |
|
146
|
|
|
Fnew = task.eval(S[i]) |
|
147
|
|
|
if (Fnew <= Fitness[i]) and (self.rand() < self.A): Sol[i], Fitness[i] = S[i], Fnew |
|
148
|
|
|
if Fnew <= f_min: best, f_min = S[i], Fnew |
|
149
|
|
|
return Sol, Fitness, {'S': S, 'Q': Q, 'v': v} |
|
150
|
|
|
|
|
151
|
|
|
# vim: tabstop=3 noexpandtab shiftwidth=3 softtabstop=3 |
|
152
|
|
|
|
NiaOrg /
NiaPy
Loading history...