|
1
|
|
|
# encoding=utf8 |
|
2
|
|
|
import logging |
|
3
|
|
|
|
|
4
|
|
|
from NiaPy.algorithms.modified import SelfAdaptiveBatAlgorithm |
|
5
|
|
|
from NiaPy.algorithms.basic.de import CrossBest1 |
|
6
|
|
|
|
|
7
|
|
|
logging.basicConfig() |
|
8
|
|
|
logger = logging.getLogger('NiaPy.algorithms.modified') |
|
9
|
|
|
logger.setLevel('INFO') |
|
10
|
|
|
|
|
11
|
|
|
__all__ = ['HybridSelfAdaptiveBatAlgorithm'] |
|
12
|
|
|
|
|
13
|
|
|
class HybridSelfAdaptiveBatAlgorithm(SelfAdaptiveBatAlgorithm): |
|
14
|
|
|
r"""Implementation of Hybrid self adaptive bat algorithm. |
|
15
|
|
|
|
|
16
|
|
|
Algorithm: |
|
17
|
|
|
Hybrid self adaptive bat algorithm |
|
18
|
|
|
|
|
19
|
|
|
Date: |
|
20
|
|
|
April 2019 |
|
21
|
|
|
|
|
22
|
|
|
Author: |
|
23
|
|
|
Klemen Berkovič |
|
24
|
|
|
|
|
25
|
|
|
License: |
|
26
|
|
|
MIT |
|
27
|
|
|
|
|
28
|
|
|
Reference paper: |
|
29
|
|
|
Fister, Iztok, Simon Fong, and Janez Brest. "A novel hybrid self-adaptive bat algorithm." The Scientific World Journal 2014 (2014). |
|
30
|
|
|
|
|
31
|
|
|
Reference URL: |
|
32
|
|
|
https://www.hindawi.com/journals/tswj/2014/709738/cta/ |
|
33
|
|
|
|
|
34
|
|
|
Attributes: |
|
35
|
|
|
Name (List[str]): List of strings representing algorithm name. |
|
36
|
|
|
F (float): Scaling factor for local search. |
|
37
|
|
|
CR (float): Probability of crossover for local search. |
|
38
|
|
|
CrossMutt (Callable[[numpy.ndarray, int, numpy.ndarray, float, float, mtrand.RandomState, Dict[str, Any]): Local search method based of Differential evolution strategy. |
|
39
|
|
|
|
|
40
|
|
|
See Also: |
|
41
|
|
|
* :class:`NiaPy.algorithms.basic.BatAlgorithm` |
|
42
|
|
|
""" |
|
43
|
|
|
Name = ['HybridSelfAdaptiveBatAlgorithm', 'HSABA'] |
|
44
|
|
|
|
|
45
|
|
|
@staticmethod |
|
46
|
|
|
def algorithmInfo(): |
|
47
|
|
|
r"""Get basic information about the algorithm. |
|
48
|
|
|
|
|
49
|
|
|
Returns: |
|
50
|
|
|
str: Basic information. |
|
51
|
|
|
""" |
|
52
|
|
|
return r"""Fister, Iztok, Simon Fong, and Janez Brest. "A novel hybrid self-adaptive bat algorithm." The Scientific World Journal 2014 (2014).""" |
|
53
|
|
|
|
|
54
|
|
|
@staticmethod |
|
55
|
|
|
def typeParameters(): |
|
56
|
|
|
r"""Get dictionary with functions for checking values of parameters. |
|
57
|
|
|
|
|
58
|
|
|
Returns: |
|
59
|
|
|
Dict[str, Callable]: Additional arguments. |
|
60
|
|
|
|
|
61
|
|
|
See Also: |
|
62
|
|
|
* :func:`NiaPy.algorithms.basic.BatAlgorithm.typeParameters` |
|
63
|
|
|
""" |
|
64
|
|
|
d = SelfAdaptiveBatAlgorithm.typeParameters() |
|
65
|
|
|
d.update({ |
|
66
|
|
|
'F': lambda x: isinstance(x, (int, float)) and x > 0, |
|
67
|
|
|
'CR': lambda x: isinstance(x, float) and 0 <= x <= 1 |
|
68
|
|
|
}) |
|
69
|
|
|
return d |
|
70
|
|
|
|
|
71
|
|
|
def setParameters(self, F=0.9, CR=0.85, CrossMutt=CrossBest1, **ukwargs): |
|
72
|
|
|
r"""Set core parameters of HybridBatAlgorithm algorithm. |
|
73
|
|
|
|
|
74
|
|
|
Arguments: |
|
75
|
|
|
F (Optional[float]): Scaling factor for local search. |
|
76
|
|
|
CR (Optional[float]): Probability of crossover for local search. |
|
77
|
|
|
CrossMutt (Optional[Callable[[numpy.ndarray, int, numpy.ndarray, float, float, mtrand.RandomState, Dict[str, Any], numpy.ndarray]]): Local search method based of Differential evolution strategy. |
|
78
|
|
|
ukwargs (Dict[str, Any]): Additional arguments. |
|
79
|
|
|
|
|
80
|
|
|
See Also: |
|
81
|
|
|
* :func:`NiaPy.algorithms.basic.BatAlgorithm.setParameters` |
|
82
|
|
|
""" |
|
83
|
|
|
SelfAdaptiveBatAlgorithm.setParameters(self, **ukwargs) |
|
84
|
|
|
self.F, self.CR, self.CrossMutt = F, CR, CrossMutt |
|
85
|
|
|
|
|
86
|
|
|
def getParameters(self): |
|
87
|
|
|
r"""Get parameters of the algorithm. |
|
88
|
|
|
|
|
89
|
|
|
Returns: |
|
90
|
|
|
Dict[str, Any]: Parameters of the algorithm. |
|
91
|
|
|
|
|
92
|
|
|
See Also: |
|
93
|
|
|
* :func:`NiaPy.algorithms.modified.AdaptiveBatAlgorithm.getParameters` |
|
94
|
|
|
""" |
|
95
|
|
|
d = SelfAdaptiveBatAlgorithm.getParameters(self) |
|
96
|
|
|
d.update({ |
|
97
|
|
|
'F': self.F, |
|
98
|
|
|
'CR': self.CR |
|
99
|
|
|
}) |
|
100
|
|
|
return d |
|
101
|
|
|
|
|
102
|
|
|
def localSearch(self, best, A, i, Sol, task, **kwargs): |
|
103
|
|
|
r"""Improve the best solution. |
|
104
|
|
|
|
|
105
|
|
|
Args: |
|
106
|
|
|
best (numpy.ndarray): Global best individual. |
|
107
|
|
|
task (Task): Optimization task. |
|
108
|
|
|
i (int): Index of current individual. |
|
109
|
|
|
Sol (numpy.ndarray): Current best population. |
|
110
|
|
|
**kwargs (Dict[str, Any]): |
|
111
|
|
|
|
|
112
|
|
|
Returns: |
|
113
|
|
|
numpy.ndarray: New solution based on global best individual. |
|
114
|
|
|
""" |
|
115
|
|
|
return task.repair(self.CrossMutt(Sol, i, best, self.F, self.CR, rnd=self.Rand), rnd=self.Rand) |
|
116
|
|
|
|
|
117
|
|
|
# vim: tabstop=3 noexpandtab shiftwidth=3 softtabstop=3 |
|
118
|
|
|
|
NiaOrg /
NiaPy
