def __init__(self,reader_obj,model_obj,option_obj):
self.fit_obj=fF(reader_obj,model_obj,option_obj)
self.SetFFun(option_obj)
self.rand=Random()
self.seed=option_obj.seed
self.rand.seed(self.seed)
self.evo_strat=ec.SA(self.rand)
self.evo_strat.terminator=terminators.evaluation_termination
if option_obj.output_level=="1":
self.evo_strat.observer=[observers.population_observer,observers.file_observer]
else:
self.evo_strat.observer=[observers.file_observer]
self.max_evaluation=option_obj.max_evaluation
self.mutation_rate=option_obj.mutation_rate
self.g_m=option_obj.m_gauss
self.g_std=option_obj.std_gauss
self.inint_T=option_obj.init_temp
self.cooling_rate=option_obj.cooling_rate
self.num_params=option_obj.num_params
self.SetBoundaries(option_obj.boundaries)
self.maximize=False #hard wired, always minimize
self.stat_file=open("stat_file.txt","w")
self.ind_file=open("ind_file.txt","w")
#inspyred needs sequence of seeds
#self.starting_points=[normalize(args.get("starting_points",uniform(self.rand,{"num_params" : self.num_params,"self": self})),self)]
try:
if isinstance(option_obj.starting_points[0],list):
self.starting_points=option_obj.starting_points
else:
self.starting_points=[normalize(option_obj.starting_points,self)]
except TypeError:
self.starting_points=None
if option_obj.output_level=="1":
print "starting points: ",self.starting_points
def __init__(self,reader_obj,model_obj,option_obj):
self.fit_obj=fF(reader_obj,model_obj,option_obj)
self.SetFFun(option_obj)
self.rand=Random()
self.seed=option_obj.seed
self.rand.seed(self.seed)
self.evo_strat=ec.SA(self.rand)
self.evo_strat.terminator=terminators.evaluation_termination
if option_obj.output_level=="1":
self.evo_strat.observer=[observers.population_observer,observers.file_observer]
else:
self.evo_strat.observer=[observers.file_observer]
self.max_evaluation=option_obj.max_evaluation
self.mutation_rate=option_obj.mutation_rate
self.g_m=option_obj.m_gauss
self.g_std=option_obj.std_gauss
self.inint_T=option_obj.init_temp
self.cooling_rate=option_obj.cooling_rate
self.num_inputs=option_obj.num_inputs
self.SetBoundaries(option_obj.boundaries)
self.maximize=False #hard wired, always minimize
self.stat_file=open("stat_file.txt","w")
self.ind_file=open("ind_file.txt","w")
#inspyred needs sequence of seeds
#self.starting_points=[normalize(args.get("starting_points",uniform(self.rand,{"num_inputs" : self.num_inputs,"self": self})),self)]
try:
if isinstance(option_obj.starting_points[0],list):
self.starting_points=option_obj.starting_points
else:
self.starting_points=[normalize(option_obj.starting_points,self)]
except TypeError:
self.starting_points=None
if option_obj.output_level=="1":
print "starting points: ",self.starting_points
def __init__(self,reader_obj,model_obj,option_obj,resolution):
self.fit_obj=fF(reader_obj,model_obj,option_obj)
self.SetFFun(option_obj)
self.num_inputs=option_obj.num_inputs
self.num_points_per_dim=resolution
#self.resolution=5
#print self.resolution
self.SetBoundaries(option_obj.boundaries)
def __init__(self,reader_obj,model_obj,option_obj,resolution):
self.fit_obj=fF(reader_obj,model_obj,option_obj)
self.SetFFun(option_obj)
self.num_params=option_obj.num_params
self.num_points_per_dim=resolution
#self.resolution=5
#print self.resolution
self.SetBoundaries(option_obj.boundaries)
def __init__(self,reader_obj,model_obj,option_obj):
self.fit_obj=fF(reader_obj,model_obj,option_obj)
self.SetFFun(option_obj)
self.rand=Random()
self.seed=option_obj.seed
self.rand.seed(self.seed)
#PSO algorithm
self.evo_strat=inspyred.swarm.PSO(self.rand)
#algorithm terminates at max number of generations
self.evo_strat.terminator=terminators.generation_termination
if option_obj.output_level=="1":
self.evo_strat.observer=[observers.population_observer,observers.file_observer]
else:
self.evo_strat.observer=[observers.file_observer]
self.max_evaluation=option_obj.max_evaluation
self.pop_size=option_obj.pop_size
#PSO attributes
self.inertia=option_obj.inertia
self.cognitive_rate=option_obj.cognitive_rate
self.social_rate=option_obj.social_rate
#self.neighborhood_size=int(round(option_obj.neighborhood_size))
self.topology=inspyred.swarm.topologies.star_topology
self.num_params=option_obj.num_params
self.SetBoundaries(option_obj.boundaries)
self.maximize=False #hard wired, always minimize
self.stat_file=open("stat_file.txt","w")
self.ind_file=open("ind_file.txt","w")
#inspyred needs sequence of seeds
#self.starting_points=[normalize(args.get("starting_points",uniform(self.rand,{"num_params" : self.num_params,"self": self})),self)]
try:
if isinstance(option_obj.starting_points[0],list):
self.starting_points=option_obj.starting_points
else:
self.starting_points=[normalize(option_obj.starting_points,self)]
except TypeError:
self.starting_points=None
if option_obj.output_level=="1":
print "starting points: ",self.starting_points
def __init__(self,reader_obj,model_obj,option_obj):
self.fit_obj=fF(reader_obj,model_obj,option_obj)
self.SetFFun(option_obj)
self.rand=Random()
self.seed=option_obj.seed
self.rand.seed(self.seed)
#PSO algorithm
self.evo_strat=inspyred.swarm.PSO(self.rand)
#algorithm terminates at max number of generations
self.evo_strat.terminator=terminators.generation_termination
if option_obj.output_level=="1":
self.evo_strat.observer=[observers.population_observer,observers.file_observer]
else:
self.evo_strat.observer=[observers.file_observer]
self.max_evaluation=option_obj.max_evaluation
self.pop_size=option_obj.pop_size
#PSO attributes
self.inertia=option_obj.inertia
self.cognitive_rate=option_obj.cognitive_rate
self.social_rate=option_obj.social_rate
#self.neighborhood_size=int(round(option_obj.neighborhood_size))
self.topology=inspyred.swarm.topologies.star_topology
self.num_params=option_obj.num_params
self.SetBoundaries(option_obj.boundaries)
self.maximize=False #hard wired, always minimize
self.stat_file=open("stat_file.txt","w")
self.ind_file=open("ind_file.txt","w")
#inspyred needs sequence of seeds
#self.starting_points=[normalize(args.get("starting_points",uniform(self.rand,{"num_params" : self.num_params,"self": self})),self)]
try:
if isinstance(option_obj.starting_points[0],list):
self.starting_points=option_obj.starting_points
else:
self.starting_points=[normalize(option_obj.starting_points,self)]
except TypeError:
self.starting_points=None
if option_obj.output_level=="1":
print "starting points: ",self.starting_points
def __init__(self,reader_obj,model_obj,option_obj):
self.fit_obj=fF(reader_obj,model_obj,option_obj)
self.SetFFun(option_obj)
self.rand=Random()
self.seed=option_obj.seed
self.rand.seed(self.seed)
self.evo_strat=ec.DEA(self.rand)
# - *num_selected* -- the number of individuals to be selected (default 2)
# - *tournament_size* -- the tournament size (default 2)
# - *crossover_rate* -- the rate at which crossover is performed
# (default 1.0)
# - *mutation_rate* -- the rate at which mutation is performed (default 0.1)
# - *gaussian_mean* -- the mean used in the Gaussian function (default 0)
# - *gaussian_stdev* -- the standard deviation used in the Gaussian function
# (default 1)
self.evo_strat.terminator=terminators.generation_termination
if option_obj.output_level=="1":
self.evo_strat.observer=[observers.population_observer,observers.file_observer]
else:
self.evo_strat.observer=[observers.file_observer]
self.pop_size=option_obj.pop_size
self.max_evaluation=option_obj.max_evaluation
self.mutation_rate=option_obj.mutation_rate
self.crossover_rate=option_obj.crossover_rate
self.num_params=option_obj.num_params
self.SetBoundaries(option_obj.boundaries)
self.maximize=False #hard wired, always minimize
self.stat_file=open("stat_file.txt","w")
self.ind_file=open("ind_file.txt","w")
#inspyred needs sequence of seeds
#self.starting_points=[normalize(args.get("starting_points",uniform(self.rand,{"num_params" : self.num_params,"self": self})),self)]
try:
#print type(option_obj.starting_points)
if isinstance(option_obj.starting_points[0],list):
self.starting_points=option_obj.starting_points
else:
self.starting_points=[normalize(option_obj.starting_points,self)]
except TypeError:
self.starting_points=None
if option_obj.output_level=="1":
print "starting points: ",self.starting_points
def __init__(self,reader_obj,model_obj,option_obj):
self.fit_obj=fF(reader_obj,model_obj,option_obj)
self.SetFFun(option_obj)
self.rand=Random()
self.seed=option_obj.seed
self.rand.seed(self.seed)
self.evo_strat=ec.DEA(self.rand)
# - *num_selected* -- the number of individuals to be selected (default 2)
# - *tournament_size* -- the tournament size (default 2)
# - *crossover_rate* -- the rate at which crossover is performed
# (default 1.0)
# - *mutation_rate* -- the rate at which mutation is performed (default 0.1)
# - *gaussian_mean* -- the mean used in the Gaussian function (default 0)
# - *gaussian_stdev* -- the standard deviation used in the Gaussian function
# (default 1)
self.evo_strat.terminator=terminators.generation_termination
if option_obj.output_level=="1":
self.evo_strat.observer=[observers.population_observer,observers.file_observer]
else:
self.evo_strat.observer=[observers.file_observer]
self.pop_size=option_obj.pop_size
self.max_evaluation=option_obj.max_evaluation
self.mutation_rate=option_obj.mutation_rate
self.crossover_rate=option_obj.crossover_rate
self.num_params=option_obj.num_params
self.SetBoundaries(option_obj.boundaries)
self.maximize=False #hard wired, always minimize
self.stat_file=open("stat_file.txt","w")
self.ind_file=open("ind_file.txt","w")
#inspyred needs sequence of seeds
#self.starting_points=[normalize(args.get("starting_points",uniform(self.rand,{"num_params" : self.num_params,"self": self})),self)]
try:
#print type(option_obj.starting_points)
if isinstance(option_obj.starting_points[0],list):
self.starting_points=option_obj.starting_points
else:
self.starting_points=[normalize(option_obj.starting_points,self)]
except TypeError:
self.starting_points=None
if option_obj.output_level=="1":
print "starting points: ",self.starting_points
def __init__(self,reader_obj,model_obj,option_obj):
self.fit_obj=fF(reader_obj,model_obj,option_obj)
self.SetFFun(option_obj)
self.rand=Random()
self.seed=option_obj.seed
self.rand.seed(self.seed)
self.pop_size=option_obj.pop_size
self.num_inputs=option_obj.num_inputs
self.SetBoundaries(option_obj.boundaries)
try:
#print type(option_obj.starting_points)
if isinstance(option_obj.starting_points[0],list):
self.starting_points=option_obj.starting_points
else:
self.starting_points=[normalize(option_obj.starting_points,self)]
except TypeError:
self.starting_points=None
if option_obj.output_level=="1":
print "starting points: ",self.starting_points
def __init__(self,reader_obj,model_obj,option_obj):
self.fit_obj=fF(reader_obj,model_obj,option_obj)
self.SetFFun(option_obj)
self.rand=random
self.seed=option_obj.seed
self.rand.seed([self.seed])
self.max_evaluation=option_obj.max_evaluation
self.accuracy=option_obj.acc
self.num_inputs=option_obj.num_inputs
self.SetBoundaries(option_obj.boundaries)
try:
if isinstance(option_obj.starting_points[0],list):
raise TypeError
else:
self.starting_points=[normalize(option_obj.starting_points,self)]
except TypeError:
self.starting_points=uniform(self.rand,{"num_inputs" : self.num_inputs,"self": self})
if option_obj.output_level=="1":
print "starting points: ",self.starting_points
def __init__(self,reader_obj,model_obj,option_obj):
self.fit_obj=fF(reader_obj,model_obj,option_obj)
self.SetFFun(option_obj)
self.rand=Random()
self.seed=option_obj.seed
self.rand.seed(self.seed)
self.pop_size=option_obj.pop_size
self.num_params=option_obj.num_params
self.SetBoundaries(option_obj.boundaries)
try:
#print type(option_obj.starting_points)
if isinstance(option_obj.starting_points[0],list):
self.starting_points=option_obj.starting_points
else:
self.starting_points=[normalize(option_obj.starting_points,self)]
except TypeError:
self.starting_points=None
if option_obj.output_level=="1":
print "starting points: ",self.starting_points
def __init__(self,reader_obj,model_obj,option_obj):
self.fit_obj=fF(reader_obj,model_obj,option_obj)
self.SetFFun(option_obj)
self.rand=random
self.seed=option_obj.seed
self.rand.seed([self.seed])
self.max_evaluation=option_obj.max_evaluation
self.accuracy=option_obj.acc
self.num_params=option_obj.num_params
self.SetBoundaries(option_obj.boundaries)
try:
if isinstance(option_obj.starting_points[0],list):
raise TypeError
else:
self.starting_points=[normalize(option_obj.starting_points,self)]
except TypeError:
self.starting_points=uniform(self.rand,{"num_params" : self.num_params,"self": self})
if option_obj.output_level=="1":
print "starting points: ",self.starting_points
def __init__(self,reader_obj,model_obj,option_obj):
self.fit_obj=fF(reader_obj,model_obj,option_obj)
self.SetFFun(option_obj)
self.rand=random
self.seed=option_obj.seed
self.rand.seed([self.seed])
self.temp=option_obj.temperature
self.num_iter=option_obj.num_iter
self.num_repet=option_obj.num_repet
self.step_size=option_obj.step_size
self.freq=option_obj.update_freq
self.num_inputs=option_obj.num_inputs
self.SetBoundaries(option_obj.boundaries)
try:
if isinstance(option_obj.starting_points[0],list):
self.starting_points=option_obj.starting_points
else:
self.starting_points=[normalize(option_obj.starting_points,self)]
except TypeError:
self.starting_points=uniform(self.rand,{"num_inputs" : self.num_inputs,"self": self})
if option_obj.output_level=="1":
print "starting points: ",self.starting_points
def __init__(self,reader_obj,model_obj,option_obj):
self.fit_obj=fF(reader_obj,model_obj,option_obj)
self.SetFFun(option_obj)
self.rand=random
self.seed=option_obj.seed
self.rand.seed([self.seed])
self.temp=option_obj.temperature
self.num_iter=option_obj.num_iter
self.num_repet=option_obj.num_repet
self.step_size=option_obj.step_size
self.freq=option_obj.update_freq
self.num_params=option_obj.num_params
self.SetBoundaries(option_obj.boundaries)
try:
if isinstance(option_obj.starting_points[0],list):
self.starting_points=option_obj.starting_points
else:
self.starting_points=[normalize(option_obj.starting_points,self)]
except TypeError:
self.starting_points=uniform(self.rand,{"num_params" : self.num_params,"self": self})
if option_obj.output_level=="1":
print "starting points: ",self.starting_points
def __init__(self,reader_obj,model_obj,option_obj):
self.fit_obj=fF(reader_obj,model_obj,option_obj)
self.SetFFun(option_obj)
self.rand=Random()
self.seed=option_obj.seed
self.rand.seed(self.seed)
self.evo_strat=ec.ES(self.rand)
self.evo_strat.terminator=terminators.generation_termination
self.evo_strat.selector=inspyred.ec.selectors.default_selection
self.evo_strat.replacer=inspyred.ec.replacers.generational_replacement
self.evo_strat.variator=[variators.gaussian_mutation,
variators.blend_crossover]
if option_obj.output_level=="1":
self.evo_strat.observer=[observers.population_observer,observers.file_observer]
else:
self.evo_strat.observer=[observers.file_observer]
self.pop_size=option_obj.pop_size
self.max_evaluation=option_obj.max_evaluation
self.mutation_rate=option_obj.mutation_rate
self.num_inputs=option_obj.num_inputs
self.SetBoundaries(option_obj.boundaries)
self.maximize=False #hard wired, always minimize
self.stat_file=open("stat_file.txt","w")
self.ind_file=open("ind_file.txt","w")
#inspyred needs sequence of seeds
#self.starting_points=[normalize(args.get("starting_points",uniform(self.rand,{"num_inputs" : self.num_inputs,"self": self})),self)]
try:
#print type(option_obj.starting_points)
if isinstance(option_obj.starting_points[0],list):
self.starting_points=option_obj.starting_points
else:
self.starting_points=[normalize(option_obj.starting_points,self)]
except TypeError:
self.starting_points=None
if option_obj.output_level=="1":
print "starting points: ",self.starting_points
def __init__(self,reader_obj,model_obj,option_obj):
self.fit_obj=fF(reader_obj,model_obj,option_obj)
self.SetFFun(option_obj)
self.rand=Random()
self.seed=option_obj.seed
self.rand.seed(self.seed)
self.evo_strat=ec.ES(self.rand)
self.evo_strat.terminator=terminators.generation_termination
self.evo_strat.selector=inspyred.ec.selectors.default_selection
self.evo_strat.replacer=inspyred.ec.replacers.generational_replacement
self.evo_strat.variator=[variators.gaussian_mutation,
variators.blend_crossover]
if option_obj.output_level=="1":
self.evo_strat.observer=[observers.population_observer,observers.file_observer]
else:
self.evo_strat.observer=[observers.file_observer]
self.pop_size=option_obj.pop_size
self.max_evaluation=option_obj.max_evaluation
self.mutation_rate=option_obj.mutation_rate
self.num_params=option_obj.num_params
self.SetBoundaries(option_obj.boundaries)
self.maximize=False #hard wired, always minimize
self.stat_file=open("stat_file.txt","w")
self.ind_file=open("ind_file.txt","w")
#inspyred needs sequence of seeds
#self.starting_points=[normalize(args.get("starting_points",uniform(self.rand,{"num_params" : self.num_params,"self": self})),self)]
try:
#print type(option_obj.starting_points)
if isinstance(option_obj.starting_points[0],list):
self.starting_points=option_obj.starting_points
else:
self.starting_points=[normalize(option_obj.starting_points,self)]
except TypeError:
self.starting_points=None
if option_obj.output_level=="1":
print "starting points: ",self.starting_points
