# encoding=utf8
import logging
from numpy import apply_along_axis, argmin, argmax, sum, sqrt, round, argsort, fabs, asarray, where
from NiaPy.algorithms.algorithm import Algorithm
from NiaPy.util import fullArray
logging.basicConfig()
logger = logging.getLogger('NiaPy.algorithms.basic')
logger.setLevel('INFO')
__all__ = ['FireworksAlgorithm', 'EnhancedFireworksAlgorithm', 'DynamicFireworksAlgorithm', 'DynamicFireworksAlgorithmGauss', 'BareBonesFireworksAlgorithm']
[docs]class BareBonesFireworksAlgorithm(Algorithm):
r"""Implementation of Bare Bones Fireworks Algorithm.
Algorithm:
Bare Bones Fireworks Algorithm
Date:
2018
Authors:
Klemen Berkovič
License:
MIT
Reference URL:
https://www.sciencedirect.com/science/article/pii/S1568494617306609
Reference paper:
Junzhi Li, Ying Tan, The bare bones fireworks algorithm: A minimalist global optimizer, Applied Soft Computing, Volume 62, 2018, Pages 454-462, ISSN 1568-4946, https://doi.org/10.1016/j.asoc.2017.10.046.
Attributes:
Name (lsit of str): List of strings representing algorithm names
n (int): Number of spraks
C_a (float): amplification coefficient
C_r (float): reduction coefficient
"""
Name = ['BareBonesFireworksAlgorithm', 'BBFWA']
[docs] @staticmethod
def algorithmInfo():
r"""Get default information of algorithm.
Returns:
str: Basic information.
See Also:
* :func:`NiaPy.algorithms.Algorithm.algorithmInfo`
"""
return r"""Junzhi Li, Ying Tan, The bare bones fireworks algorithm: A minimalist global optimizer, Applied Soft Computing, Volume 62, 2018, Pages 454-462, ISSN 1568-4946, https://doi.org/10.1016/j.asoc.2017.10.046."""
[docs] @staticmethod
def typeParameters(): return {
'n': lambda x: isinstance(x, int) and x > 0,
'C_a': lambda x: isinstance(x, (float, int)) and x > 1,
'C_r': lambda x: isinstance(x, (float, int)) and 0 < x < 1
}
[docs] def setParameters(self, n=10, C_a=1.5, C_r=0.5, **ukwargs):
r"""Set the arguments of an algorithm.
Arguments:
n (int): Number of sparks :math:`\in [1, \infty)`.
C_a (float): Amplification coefficient :math:`\in [1, \infty)`.
C_r (float): Reduction coefficient :math:`\in (0, 1)`.
"""
ukwargs.pop('NP', None)
Algorithm.setParameters(self, NP=1, **ukwargs)
self.n, self.C_a, self.C_r = n, C_a, C_r
[docs] def initPopulation(self, task):
r"""Initialize starting population.
Args:
task (Task): Optimization task.
Returns:
Tuple[numpy.ndarray, float, Dict[str, Any]]:
1. Initial solution.
2. Initial solution function/fitness value.
3. Additional arguments:
* A (numpy.ndarray): Starting aplitude or search range.
"""
x, x_fit, d = Algorithm.initPopulation(self, task)
d.update({'A': task.bRange})
return x, x_fit, d
[docs] def runIteration(self, task, x, x_fit, xb, fxb, A, **dparams):
r"""Core function of Bare Bones Fireworks Algorithm.
Args:
task (Task): Optimization task.
x (numpy.ndarray): Current solution.
x_fit (float): Current solution fitness/function value.
xb (numpy.ndarray): Current best solution.
fxb (float): Current best solution fitness/function value.
A (numpy.ndarray): Serach range.
dparams (Dict[str, Any]): Additional parameters.
Returns:
Tuple[numpy.ndarray, float, numpy.ndarray, float, Dict[str, Any]]:
1. New solution.
2. New solution fitness/function value.
3. New global best solution.
4. New global best solutions fitness/objective value.
5. Additional arguments:
* A (numpy.ndarray): Serach range.
"""
S = apply_along_axis(task.repair, 1, self.uniform(x - A, x + A, [self.n, task.D]), self.Rand)
S_fit = apply_along_axis(task.eval, 1, S)
iS = argmin(S_fit)
if S_fit[iS] < x_fit: x, x_fit, A = S[iS], S_fit[iS], self.C_a * A
else: A = self.C_r * A
return x, x_fit, x.copy(), x_fit, {'A': A}
[docs]class FireworksAlgorithm(Algorithm):
r"""Implementation of fireworks algorithm.
Algorithm:
Fireworks Algorithm
Date:
2018
Authors:
Klemen Berkovič
License:
MIT
Reference URL:
https://www.springer.com/gp/book/9783662463529
Reference paper:
Tan, Ying. "Fireworks algorithm." Heidelberg, Germany: Springer 10 (2015): 978-3
Attributes:
Name (List[str]): List of stirngs representing algorithm names.
"""
Name = ['FireworksAlgorithm', 'FWA']
[docs] @staticmethod
def algorithmInfo():
r"""Get default information of algorithm.
Returns:
str: Basic information.
See Also:
* :func:`NiaPy.algorithms.Algorithm.algorithmInfo`
"""
return r"""Tan, Ying. "Fireworks algorithm." Heidelberg, Germany: Springer 10 (2015): 978-3."""
[docs] @staticmethod
def typeParameters(): return {
'N': lambda x: isinstance(x, int) and x > 0,
'm': lambda x: isinstance(x, int) and x > 0,
'a': lambda x: isinstance(x, (int, float)) and x > 0,
'b': lambda x: isinstance(x, (int, float)) and x > 0,
'epsilon': lambda x: isinstance(x, float) and 0 < x < 1
}
[docs] def setParameters(self, N=40, m=40, a=1, b=2, A=40, epsilon=1e-31, **ukwargs):
r"""Set the arguments of an algorithm.
Arguments:
N (int): Number of Fireworks
m (int): Number of sparks
a (int): Limitation of sparks
b (int): Limitation of sparks
A (float): --
epsilon (float): Small number for non 0 devision
"""
Algorithm.setParameters(self, NP=N, **ukwargs)
self.m, self.a, self.b, self.A, self.epsilon = m, a, b, A, epsilon
[docs] def initAmplitude(self, task):
r"""Initialize amplitudes for dimensions.
Args:
task (Task): Optimization task.
Returns:
numpy.ndarray[float]: Starting amplitudes.
"""
return fullArray(self.A, task.D)
[docs] def SparsksNo(self, x_f, xw_f, Ss):
r"""Calculate number of sparks based on function value of individual.
Args:
x_f (float): Individuals function/fitness value.
xw_f (float): Worst individual function/fitness value.
Ss (): TODO
Returns:
int: Number of sparks that individual will create.
"""
s = self.m * (xw_f - x_f + self.epsilon) / (Ss + self.epsilon)
return round(self.b * self.m) if s > self.b * self.m and self.a < self.b < 1 else round(self.a * self.m)
[docs] def ExplosionAmplitude(self, x_f, xb_f, A, As):
r"""Calculate explosion amplitude.
Args:
x_f (float): Individuals function/fitness value.
xb_f (float): Best individuals function/fitness value.
A (numpy.ndarray): Amplitudes.
As ():
Returns:
numpy.ndarray: TODO.
"""
return A * (x_f - xb_f - self.epsilon) / (As + self.epsilon)
[docs] def ExplodeSpark(self, x, A, task):
r"""Explode a spark.
Args:
x (numpy.ndarray): Individuals creating spark.
A (numpy.ndarray): Amplitude of spark.
task (Task): Optimization task.
Returns:
numpy.ndarray: Sparks exploded in with specified amplitude.
"""
return self.Mapping(x + self.rand(task.D) * self.uniform(-A, A, task.D), task)
[docs] def GaussianSpark(self, x, task):
r"""Create gaussian spark.
Args:
x (numpy.ndarray): Individual creating a spark.
task (Task): Optimization task.
Returns:
numpy.ndarray: Spark exploded based on gaussian amplitude.
"""
return self.Mapping(x + self.rand(task.D) * self.normal(1, 1, task.D), task)
[docs] def Mapping(self, x, task):
r"""Fix value to bounds..
Args:
x (numpy.ndarray): Individual to fix.
task (Task): Optimization task.
Returns:
numpy.ndarray: Individual in search range.
"""
ir = where(x > task.Upper)
x[ir] = task.Lower[ir] + x[ir] % task.bRange[ir]
ir = where(x < task.Lower)
x[ir] = task.Lower[ir] + x[ir] % task.bRange[ir]
return x
[docs] def R(self, x, FW):
r"""Calculate ranges.
Args:
x (numpy.ndarray): Individual in population.
FW (numpy.ndarray): Current population.
Returns:
numpy,ndarray[float]: Ranges values.
"""
return sqrt(sum(fabs(x - FW)))
[docs] def p(self, r, Rs):
r"""Calculate p.
Args:
r (float): Range of individual.
Rs (float): Sum of ranges.
Returns:
float: p value.
"""
return r / Rs
[docs] def NextGeneration(self, FW, FW_f, FWn, task):
r"""Generate new generation of individuals.
Args:
FW (numpy.ndarray): Current population.
FW_f (numpy.ndarray[float]): Currents population fitness/function values.
FWn (numpy.ndarray): New population.
task (Task): Optimization task.
Returns:
Tuple[numpy.ndarray, numpy.ndarray[float]]:
1. New population.
2. New populations fitness/function values.
"""
FWn_f = apply_along_axis(task.eval, 1, FWn)
ib = argmin(FWn_f)
if FWn_f[ib] < FW_f[0]: FW[0], FW_f[0] = FWn[ib], FWn_f[ib]
R = asarray([self.R(FWn[i], FWn) for i in range(len(FWn))])
Rs = sum(R)
P = asarray([self.p(R[i], Rs) for i in range(len(FWn))])
isort = argsort(P)[-(self.NP - 1):]
FW[1:], FW_f[1:] = asarray(FWn)[isort], FWn_f[isort]
return FW, FW_f
[docs] def initPopulation(self, task):
r"""Initialize starting population.
Args:
task (Task): Optimization task.
Returns:
Tuple[numpy.ndarray, numpy.ndarray[float], Dict[str, Any]]:
1. Initialized population.
2. Initialized populations function/fitness values.
3. Additional arguments:
* Ah (numpy.ndarray): Initialized amplitudes.
See Also:
* :func:`NiaPy.algorithms.algorithm.Algorithm.initPopulation`
"""
FW, FW_f, d = Algorithm.initPopulation(self, task)
Ah = self.initAmplitude(task)
d.update({'Ah': Ah})
return FW, FW_f, d
[docs] def runIteration(self, task, FW, FW_f, xb, fxb, Ah, **dparams):
r"""Core function of Fireworks algorithm.
Args:
task (Task): Optimization task.
FW (numpy.ndarray): Current population.
FW_f (numpy.ndarray[float]): Current populations function/fitness values.
xb (numpy.ndarray): Global best individual.
fxb (float): Global best individuals fitness/function value.
Ah (numpy.ndarray): Current amplitudes.
**dparams (Dict[str, Any)]: Additional arguments
Returns:
Tuple[numpy.ndarray, numpy.ndarray, numpy.ndarray, float, Dict[str, Any]]:
1. Initialized population.
2. Initialized populations function/fitness values.
3. New global best solution.
4. New global best solutions fitness/objective value.
5. Additional arguments:
* Ah (numpy.ndarray): Initialized amplitudes.
See Also:
* :func:`FireworksAlgorithm.SparsksNo`.
* :func:`FireworksAlgorithm.ExplosionAmplitude`
* :func:`FireworksAlgorithm.ExplodeSpark`
* :func:`FireworksAlgorithm.GaussianSpark`
* :func:`FireworksAlgorithm.NextGeneration`
"""
iw, ib = argmax(FW_f), 0
Ss, As = sum(FW_f[iw] - FW_f), sum(FW_f - FW_f[ib])
S = [self.SparsksNo(FW_f[i], FW_f[iw], Ss) for i in range(self.NP)]
A = [self.ExplosionAmplitude(FW_f[i], FW_f[ib], Ah, As) for i in range(self.NP)]
FWn = [self.ExplodeSpark(FW[i], A[i], task) for i in range(self.NP) for _ in range(S[i])]
for i in range(self.m): FWn.append(self.GaussianSpark(self.randint(self.NP), task))
FW, FW_f = self.NextGeneration(FW, FW_f, FWn, task)
xb, fxb = self.getBest(FW, FW_f, xb, fxb)
return FW, FW_f, xb, fxb, {'Ah': Ah}
[docs]class EnhancedFireworksAlgorithm(FireworksAlgorithm):
r"""Implementation of enganced fireworks algorithm.
Algorithm:
Enhanced Fireworks Algorithm
Date:
2018
Authors:
Klemen Berkovič
License:
MIT
Reference URL:
https://ieeexplore.ieee.org/document/6557813/
Reference paper:
S. Zheng, A. Janecek and Y. Tan, "Enhanced Fireworks Algorithm," 2013 IEEE Congress on Evolutionary Computation, Cancun, 2013, pp. 2069-2077. doi: 10.1109/CEC.2013.6557813
Attributes:
Name (List[str]): List of strings representing algorithm names.
Ainit (float): Initial amplitude of sparks.
Afinal (float): Maximal amplitude of sparks.
"""
Name = ['EnhancedFireworksAlgorithm', 'EFWA']
[docs] @staticmethod
def algorithmInfo():
r"""Get default information of algorithm.
Returns:
str: Basic information.
See Also:
* :func:`NiaPy.algorithms.Algorithm.algorithmInfo`
"""
return r"""S. Zheng, A. Janecek and Y. Tan, "Enhanced Fireworks Algorithm," 2013 IEEE Congress on Evolutionary Computation, Cancun, 2013, pp. 2069-2077. doi: 10.1109/CEC.2013.6557813"""
[docs] @staticmethod
def typeParameters():
r"""Get dictionary with functions for checking values of parameters.
Returns:
Dict[str, Callable]:
* Ainit (Callable[[Union[int, float]], bool]): TODO
* Afinal (Callable[[Union[int, float]], bool]): TODO
See Also:
* :func:`FireworksAlgorithm.typeParameters`
"""
d = FireworksAlgorithm.typeParameters()
d['Ainit'] = lambda x: isinstance(x, (float, int)) and x > 0
d['Afinal'] = lambda x: isinstance(x, (float, int)) and x > 0
return d
[docs] def setParameters(self, Ainit=20, Afinal=5, **ukwargs):
r"""Set EnhancedFireworksAlgorithm algorithms core parameters.
Args:
Ainit (float): TODO
Afinal (float): TODO
**ukwargs (Dict[str, Any]): Additional arguments.
See Also:
* :func:`FireworksAlgorithm.setParameters`
"""
FireworksAlgorithm.setParameters(self, **ukwargs)
self.Ainit, self.Afinal = Ainit, Afinal
[docs] def initRanges(self, task):
r"""Initialize ranges.
Args:
task (Task): Optimization task.
Returns:
Tuple[numpy.ndarray[float], numpy.ndarray[float], numpy.ndarray[float]]:
1. Initial amplitude values over dimensions.
2. Final amplitude values over dimensions.
3. uAmin.
"""
Ainit, Afinal = fullArray(self.Ainit, task.D), fullArray(self.Afinal, task.D)
return Ainit, Afinal, self.uAmin(Ainit, Afinal, task)
[docs] def uAmin(self, Ainit, Afinal, task):
r"""Calculate the value of `uAmin`.
Args:
Ainit (numpy.ndarray[float]): Initial amplitude values over dimensions.
Afinal (numpy.ndarray[float]): Final amplitude values over dimensions.
task (Task): Optimization task.
Returns:
numpy.ndarray[float]: uAmin.
"""
return Ainit - sqrt(task.Evals * (2 * task.nFES - task.Evals)) * (Ainit - Afinal) / task.nFES
[docs] def ExplosionAmplitude(self, x_f, xb_f, Ah, As, A_min=None):
r"""Calculate explosion amplitude.
Args:
x_f (float): Individuals function/fitness value.
xb_f (float): Best individual function/fitness value.
Ah (numpy.ndarray):
As (): TODO.
A_min (Optional[numpy.ndarray]): Minimal amplitude values.
task (Task): Optimization task.
Returns:
numpy.ndarray: New amplitude.
"""
A = FireworksAlgorithm.ExplosionAmplitude(self, x_f, xb_f, Ah, As)
ifix = where(A < A_min)
A[ifix] = A_min[ifix]
return A
[docs] def GaussianSpark(self, x, xb, task):
r"""Create new individual.
Args:
x (numpy.ndarray):
xb (numpy.ndarray):
task (Task): Optimization task.
Returns:
numpy.ndarray: New individual generated by gaussian noise.
"""
return self.Mapping(x + self.rand(task.D) * (xb - x) * self.normal(1, 1, task.D), task)
[docs] def NextGeneration(self, FW, FW_f, FWn, task):
r"""Generate new population.
Args:
FW (numpy.ndarray): Current population.
FW_f (numpy.ndarray[float]): Current populations fitness/function values.
FWn (numpy.ndarray): New population.
task (Task): Optimization task.
Returns:
Tuple[numpy.ndarray, numpy.ndarray[float]]:
1. New population.
2. New populations fitness/function values.
"""
FWn_f = apply_along_axis(task.eval, 1, FWn)
ib = argmin(FWn_f)
if FWn_f[ib] < FW_f[0]: FW[0], FW_f[0] = FWn[ib], FWn_f[ib]
for i in range(1, self.NP):
r = self.randint(len(FWn))
if FWn_f[r] < FW_f[i]: FW[i], FW_f[i] = FWn[r], FWn_f[r]
return FW, FW_f
[docs] def initPopulation(self, task):
r"""Initialize population.
Args:
task (Task): Optimization task.
Returns:
Tuple[numpy.ndarray, numpy.ndarray[float], Dict[str, Any]]:
1. Initial population.
2. Initial populations fitness/function values.
3. Additional arguments:
* Ainit (numpy.ndarray): Initial amplitude values.
* Afinal (numpy.ndarray): Final amplitude values.
* A_min (numpy.ndarray): Minimal amplitude values.
See Also:
* :func:`FireworksAlgorithm.initPopulation`
"""
FW, FW_f, d = FireworksAlgorithm.initPopulation(self, task)
Ainit, Afinal, A_min = self.initRanges(task)
d.update({'Ainit': Ainit, 'Afinal': Afinal, 'A_min': A_min})
return FW, FW_f, d
[docs] def runIteration(self, task, FW, FW_f, xb, fxb, Ah, Ainit, Afinal, A_min, **dparams):
r"""Core function of EnhancedFireworksAlgorithm algorithm.
Args:
task (Task): Optimization task.
FW (numpy.ndarray): Current population.
FW_f (numpy.ndarray[float]): Current populations fitness/function values.
xb (numpy.ndarray): Global best individual.
fxb (float): Global best individuals function/fitness value.
Ah (numpy.ndarray[float]): Current amplitude.
Ainit (numpy.ndarray[float]): Initial amplitude.
Afinal (numpy.ndarray[float]): Final amplitude values.
A_min (numpy.ndarray[float]): Minial amplitude values.
**dparams (Dict[str, Any]): Additional arguments.
Returns:
Tuple[numpy.ndarray, numpy.ndarray, numpy.ndarray, float, Dict[str, Any]]:
1. Initial population.
2. Initial populations fitness/function values.
3. New global best solution.
4. New global best solutions fitness/objective value.
5. Additional arguments:
* Ainit (numpy.ndarray): Initial amplitude values.
* Afinal (numpy.ndarray): Final amplitude values.
* A_min (numpy.ndarray): Minimal amplitude values.
"""
iw, ib = argmax(FW_f), 0
Ss, As = sum(FW_f[iw] - FW_f), sum(FW_f - FW_f[ib])
S = [self.SparsksNo(FW_f[i], FW_f[iw], Ss) for i in range(self.NP)]
A = [self.ExplosionAmplitude(FW_f[i], FW_f[ib], Ah, As, A_min) for i in range(self.NP)]
A_min = self.uAmin(Ainit, Afinal, task)
FWn = [self.ExplodeSpark(FW[i], A[i], task) for i in range(self.NP) for _ in range(S[i])]
for i in range(self.m): FWn.append(self.GaussianSpark(self.randint(self.NP), FW[ib], task))
FW, FW_f = self.NextGeneration(FW, FW_f, FWn, task)
xb, fxb = self.getBest(FW, FW_f, xb, fxb)
return FW, FW_f, xb, fxb, {'Ah': Ah, 'Ainit': Ainit, 'Afinal': Afinal, 'A_min': A_min}
[docs]class DynamicFireworksAlgorithmGauss(EnhancedFireworksAlgorithm):
r"""Implementation of dynamic fireworks algorithm.
Algorithm:
Dynamic Fireworks Algorithm
Date:
2018
Authors:
Klemen Berkovič
License:
MIT
Reference URL:
http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6900485&isnumber=6900223
Reference paper:
S. Zheng, A. Janecek, J. Li and Y. Tan, "Dynamic search in fireworks algorithm," 2014 IEEE Congress on Evolutionary Computation (CEC), Beijing, 2014, pp. 3222-3229. doi: 10.1109/CEC.2014.6900485
Attributes:
Name (List[str]): List of strings representing algorithm names.
A_cf (Union[float, int]): TODO
C_a (Union[float, int]): Amplification factor.
C_r (Union[float, int]): Reduction factor.
epsilon (Union[float, int]): Small value.
"""
Name = ['DynamicFireworksAlgorithmGauss', 'dynFWAG']
[docs] @staticmethod
def algorithmInfo():
r"""Get default information of algorithm.
Returns:
str: Basic information.
See Also:
* :func:`NiaPy.algorithms.Algorithm.algorithmInfo`
"""
return r"""S. Zheng, A. Janecek, J. Li and Y. Tan, "Dynamic search in fireworks algorithm," 2014 IEEE Congress on Evolutionary Computation (CEC), Beijing, 2014, pp. 3222-3229. doi: 10.1109/CEC.2014.6900485"""
[docs] @staticmethod
def typeParameters():
r"""Get dictionary with functions for checking values of parameters.
Returns:
Dict[str, Callable]:
* A_cr (Callable[[Union[float, int], bool]): TODo
See Also:
* :func:`FireworksAlgorithm.typeParameters`
"""
d = FireworksAlgorithm.typeParameters()
d['A_cf'] = lambda x: isinstance(x, (float, int)) and x > 0
d['C_a'] = lambda x: isinstance(x, (float, int)) and x > 1
d['C_r'] = lambda x: isinstance(x, (float, int)) and 0 < x < 1
d['epsilon'] = lambda x: isinstance(x, (float, int)) and 0 < x < 1
return d
[docs] def setParameters(self, A_cf=20, C_a=1.2, C_r=0.9, epsilon=1e-8, **ukwargs):
r"""Set core arguments of DynamicFireworksAlgorithmGauss.
Args:
A_cf (Union[int, float]):
C_a (Union[int, float]):
C_r (Union[int, float]):
epsilon (Union[int, float]):
**ukwargs (Dict[str, Any]): Additional arguments.
See Also:
* :func:`FireworksAlgorithm.setParameters`
"""
FireworksAlgorithm.setParameters(self, **ukwargs)
self.A_cf, self.C_a, self.C_r, self.epsilon = A_cf, C_a, C_r, epsilon
[docs] def initAmplitude(self, task):
r"""Initialize amplitude.
Args:
task (Task): Optimization task.
Returns:
Tuple[numpy.ndarray, numpy.ndarray]:
1. Initial amplitudes.
2. Amplitude for best spark.
"""
return FireworksAlgorithm.initAmplitude(self, task), task.bRange
[docs] def Mapping(self, x, task):
r"""Fix out of bound solution/individual.
Args:
x (numpy.ndarray): Individual.
task (Task): Optimization task.
Returns:
numpy.ndarray: Fixed individual.
"""
ir = where(x > task.Upper)
x[ir] = self.uniform(task.Lower[ir], task.Upper[ir])
ir = where(x < task.Lower)
x[ir] = self.uniform(task.Lower[ir], task.Upper[ir])
return x
[docs] def repair(self, x, d, epsilon):
r"""Repair solution.
Args:
x (numpy.ndarray): Individual.
d (numpy.ndarray): Default value.
epsilon (float): Limiting value.
Returns:
numpy.ndarray: Fixed solution.
"""
ir = where(x <= epsilon)
x[ir] = d[ir]
return x
[docs] def NextGeneration(self, FW, FW_f, FWn, task):
r"""TODO.
Args:
FW (numpy.ndarray): Current population.
FW_f (numpy.ndarray[float]): Current populations function/fitness values.
FWn (numpy.ndarray): New population.
task (Task): Optimization task.
Returns:
Tuple[numpy.ndarray, numpy.ndarray[float]]:
1. New population.
2. New populations function/fitness values.
"""
FWn_f = apply_along_axis(task.eval, 1, FWn)
ib = argmin(FWn_f)
for i, f in enumerate(FW_f):
r = self.randint(len(FWn))
if FWn_f[r] < f: FW[i], FW_f[i] = FWn[r], FWn_f[r]
FW[0], FW_f[0] = FWn[ib], FWn_f[ib]
return FW, FW_f
[docs] def uCF(self, xnb, xcb, xcb_f, xb, xb_f, Acf, task):
r"""TODO.
Args:
xnb:
xcb:
xcb_f:
xb:
xb_f:
Acf:
task (Task): Optimization task.
Returns:
Tuple[numpy.ndarray, float, numpy.ndarray]:
1. TODO
"""
xnb_f = apply_along_axis(task.eval, 1, xnb)
ib_f = argmin(xnb_f)
if xnb_f[ib_f] <= xb_f: xb, xb_f = xnb[ib_f], xnb_f[ib_f]
Acf = self.repair(Acf, task.bRange, self.epsilon)
if xb_f >= xcb_f: xb, xb_f, Acf = xcb, xcb_f, Acf * self.C_a
else: Acf = Acf * self.C_r
return xb, xb_f, Acf
[docs] def ExplosionAmplitude(self, x_f, xb_f, Ah, As, A_min=None):
return FireworksAlgorithm.ExplosionAmplitude(self, x_f, xb_f, Ah, As)
[docs] def initPopulation(self, task):
r"""Initialize population.
Args:
task (Task): Optimization task.
Returns:
Tuple[numpy.ndarray, numpy.ndarray[float], Dict[str, Any]]:
1. Initialized population.
2. Initialized population function/fitness values.
3. Additional arguments:
* Ah (): TODO
* Ab (): TODO
"""
FW, FW_f, _ = Algorithm.initPopulation(self, task)
Ah, Ab = self.initAmplitude(task)
return FW, FW_f, {'Ah': Ah, 'Ab': Ab}
[docs] def runIteration(self, task, FW, FW_f, xb, fxb, Ah, Ab, **dparams):
r"""Core function of DynamicFireworksAlgorithmGauss algorithm.
Args:
task (Task): Optimization task.
FW (numpy.ndarray): Current population.
FW_f (numpy.ndarray): Current populations function/fitness values.
xb (numpy.ndarray): Global best individual.
fxb (float): Global best fitness/function value.
Ah (Union[numpy.ndarray, float]): TODO
Ab (Union[numpy.ndarray, float]): TODO
**dparams (Dict[str, Any]): Additional arguments.
Returns:
Tuple[numpy.ndarray, numpy.ndarray, numpy.ndarray, float, Dict[str, Any]]:
1. New population.
2. New populations fitness/function values.
3. New global best solution.
4. New global best solutions fitness/objective value.
5. Additional arguments:
* Ah (Union[numpy.ndarray, float]): TODO
* Ab (Union[numpy.ndarray, float]): TODO
"""
iw, ib = argmax(FW_f), argmin(FW_f)
Ss, As = sum(FW_f[iw] - FW_f), sum(FW_f - FW_f[ib])
S, sb = [self.SparsksNo(FW_f[i], FW_f[iw], Ss) for i in range(len(FW))], self.SparsksNo(fxb, FW_f[iw], Ss)
A = [self.ExplosionAmplitude(FW_f[i], FW_f[ib], Ah, As) for i in range(len(FW))]
FWn, xnb = [self.ExplodeSpark(FW[i], A[i], task) for i in range(self.NP) for _ in range(S[i])], [self.ExplodeSpark(xb, Ab, task) for _ in range(sb)]
for i in range(self.m): FWn.append(self.GaussianSpark(self.randint(self.NP), FW[ib], task))
FW, FW_f = self.NextGeneration(FW, FW_f, FWn, task)
iw, ib = argmax(FW_f), 0
xb, fxb, Ab = self.uCF(xnb, FW[ib], FW_f[ib], xb, fxb, Ab, task)
return FW, FW_f, xb, fxb, {'Ah': Ah, 'Ab': Ab}
[docs]class DynamicFireworksAlgorithm(DynamicFireworksAlgorithmGauss):
r"""Implementation of dynamic fireworks algorithm.
Algorithm:
Dynamic Fireworks Algorithm
Date:
2018
Authors:
Klemen Berkovič
License:
MIT
Reference URL:
http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6900485&isnumber=6900223
Reference paper:
S. Zheng, A. Janecek, J. Li and Y. Tan, "Dynamic search in fireworks algorithm," 2014 IEEE Congress on Evolutionary Computation (CEC), Beijing, 2014, pp. 3222-3229. doi: 10.1109/CEC.2014.6900485
Attributes:
Name (List[str]): List of strings representing algorithm name.
See Also:
* :class:`NiaPy.algorithms.basic.DynamicFireworksAlgorithmGauss`
"""
Name = ['DynamicFireworksAlgorithm', 'dynFWA']
[docs] @staticmethod
def algorithmInfo():
r"""Get default information of algorithm.
Returns:
str: Basic information.
See Also:
* :func:`NiaPy.algorithms.Algorithm.algorithmInfo`
"""
return r"""S. Zheng, A. Janecek, J. Li and Y. Tan, "Dynamic search in fireworks algorithm," 2014 IEEE Congress on Evolutionary Computation (CEC), Beijing, 2014, pp. 3222-3229. doi: 10.1109/CEC.2014.6900485"""
[docs] def runIteration(self, task, FW, FW_f, xb, fxb, Ah, Ab, **dparams):
r"""Core function of Dynamic Fireworks Algorithm.
Args:
task (Task): Optimization task
FW (numpy.ndarray): Current population
FW_f (numpy.ndarray[float]): Current population fitness/function values
xb (numpy.ndarray): Current best solution
fxb (float): Current best solution's fitness/function value
Ah (): TODO
Ab (): TODO
**dparams:
Returns:
Tuple[numpy.ndarray, numpy.ndarray[float], Dict[str, Any]]:
1. New population.
2. New population function/fitness values.
3. Additional arguments:
* Ah (): TODO
* Ab (): TODO
"""
iw, ib = argmax(FW_f), argmin(FW_f)
Ss, As = sum(FW_f[iw] - FW_f), sum(FW_f - FW_f[ib])
S, sb = [self.SparsksNo(FW_f[i], FW_f[iw], Ss) for i in range(len(FW))], self.SparsksNo(fxb, FW_f[iw], Ss)
A = [self.ExplosionAmplitude(FW_f[i], FW_f[ib], Ah, As) for i in range(len(FW))]
FWn, xnb = [self.ExplodeSpark(FW[i], A[i], task) for i in range(self.NP) for _ in range(S[i])], [self.ExplodeSpark(xb, Ab, task) for _ in range(sb)]
FW, FW_f = self.NextGeneration(FW, FW_f, FWn, task)
iw, ib = argmax(FW_f), 0
xb, fxb, Ab = self.uCF(xnb, FW[ib], FW_f[ib], xb, fxb, Ab, task)
return FW, FW_f, xb, fxb, {'Ah': Ah, 'Ab': Ab}
# vim: tabstop=3 noexpandtab shiftwidth=3 softtabstop=3
