def main_run():
global cm, coords
# load the tsp data file
filehandle = open("instancias.txt", "r+")
coords = read_file(filehandle)
cm = cartesian_matrix(coords)
# set the alleles to the cities numbers
setOfAlleles = GAllele.GAlleles(homogeneous=True)
lst = [i for i in xrange(len(coords))]
a = GAllele.GAlleleList(lst)
setOfAlleles.add(a)
genome = G1DList.G1DList(len(coords))
genome.setParams(allele=setOfAlleles)
genome.evaluator.set(eval_func)
genome.mutator.set(Mutators.G1DListMutatorSwap)
genome.crossover.set(Crossovers.G1DListCrossoverOX)
genome.initializator.set(G1DListTSPInitializator)
ga = GSimpleGA.GSimpleGA(genome)
ga.setGenerations(1000)
ga.setMinimax(Consts.minimaxType["minimize"])
ga.setCrossoverRate(1.0)
ga.setMutationRate(0.03)
ga.setPopulationSize(80)
ga.evolve(freq_stats=100)
best = ga.bestIndividual()
write_tour_to_img(coords, best, "tsp_result.png")
def _make_alleles(self):
""" Returns a GAllelle.Galleles instance with alleles corresponding to
the parameters specified by the user"""
alleles = GAllele.GAlleles()
count = 0
for param in self.get_parameters().values():
allele = None
count += 1
val = param.evaluate() #now grab the value
low = param.low
high = param.high
metadata = param.get_metadata()[0][1]
#then it's a float or an int, or a member of an array
if ('low' in metadata or 'high' in metadata) or array_test.search(param.targets[0]):
if isinstance(val, real_types):
#some kind of float
allele = GAllele.GAlleleRange(begin=low, end=high, real=True)
#some kind of int
if isinstance(val, int_types):
allele = GAllele.GAlleleRange(begin=low, end=high, real=False)
elif "values" in metadata and isinstance(metadata['values'], iterable_types):
allele = GAllele.GAlleleList(metadata['values'])
if allele:
alleles.add(allele)
else:
self.raise_exception("%s is not a float, int, or enumerated \
datatype. Only these 3 types are allowed"%(param.targets[0]),ValueError)
self.count = count
return alleles
def test_createAlleleRange(self):
_allelerange = GAllele.GAlleleRange(10, 20)
self.assertEqual(_allelerange.beginEnd, [(10, 20)])
self.assertEqual(_allelerange.minimum, 10)
self.assertEqual(_allelerange.maximum, 20)
_allelerange = GAllele.GAlleleRange(1.0, 2.0, real=True)
self.assertEqual(_allelerange.real, True)
def new_genome(results_path, **kwargs):
'''
*new_genome : takes a set of alleles [function_alleles, leaves_alleles]
and options that define initializator, mutator, crossover, evaluator
and returns a genome with those options
possible initializator : "grow" : "grow" algorithm of network = recursive and possibly incomplete
possible mutator : "simple" : change genomic alleles with possible alleles with probability pmut
possible crossover :
possible evaluator : "degree_distribution", "2distributions"
possible network-type : "directed_weighted", "directed_unweighted", "undirected_weighted", "undirected_unweighted"
possible tree_type : "with_constants"
'''
evaluation_method = kwargs.get("evaluation_method")
network_type = kwargs.get("network_type")
data_path = kwargs.get("data_path")
name = kwargs.get("name")
dynamic = kwargs.get("dynamic")
extension = kwargs.get("extension")
choices = emo.get_alleles(evaluation_method, network_type)
genome = py.GTree.GTree()
#genome.setParams(nb_nodes=ne.get_number_of_nodes(results_path))
#genome.setParams(nb_edges=ne.get_number_of_edges(results_path))
genome.setParams(data_path=data_path)
genome.setParams(results_path=results_path)
genome.setParams(name=name)
genome.setParams(extension=extension)
#defines alleles : one array containing possible leaves and one containing possible functions
alleles = gall.GAlleles()
lst = gall.GAlleleList(choices)
alleles.add(lst)
genome.setParams(allele=alleles)
#defines the way to construct a random tree
genome.setParams(max_depth=int(kwargs.get("max_depth", "3")))
genome.setParams(max_siblings=int(kwargs.get("max_siblings", "2")))
genome.setParams(tree_type=kwargs.get("tree_type", "default"))
genome.initializator.set(tree_init)
#defines the how to evaluate a genome
genome.setParams(evaluation_method=evaluation_method)
if dynamic:
genome.evaluator.set(eval_func_dynamic)
else:
genome.evaluator.set(eval_func)
#defines the crossover function - default now
#defines the function that mutates trees
genome.mutator.set(mutate_tree)
#defines the network_type
genome.setParams(network_type=network_type)
#tree_init(genome)
return genome
def solve():
setOfAlleles = GAllele.GAlleles(homogeneous=True)
lst = [i for i in xrange(len(points))]
a = GAllele.GAlleleList(lst)
setOfAlleles.add(a)
genome = G1DList.G1DList(len(points))
genome.setParams(allele=setOfAlleles)
genome.evaluator.set(eval_func)
genome.mutator.set(Mutators.G1DListMutatorSwap)
genome.crossover.set(Crossovers.G1DListCrossoverOX)
genome.initializator.set(tsp_initializator)
ga = GSimpleGA.GSimpleGA(genome)
ga.setGenerations(1000)
ga.setMinimax(Consts.minimaxType["minimize"])
ga.setCrossoverRate(1.0)
ga.setMutationRate(0.03)
ga.setPopulationSize(80)
ga.evolve(freq_stats=100)
best = ga.bestIndividual()
best.append(best[0])
for i in range(len(best) - 1):
canvas.create_line(points[best[i]].x, points[best[i]].y,
points[best[i + 1]].x, points[best[i + 1]].y)
def run_main():
# Genome instance
setOfAlleles = GAllele.GAlleles()
for i in xrange(1000):
# You can even add objects instead of strings or
# primitive values
a = GAllele.GAlleleList(['UP', 'DOWN', 'LEFT', 'RIGHT'])
setOfAlleles.add(a)
# Genome instance
genome = G1DList.G1DList(300)
#genome.setParams(rangemin=0, rangemax=10)
genome.setParams(allele=setOfAlleles)
# The evaluator function (objective function)
genome.evaluator.set(eval_func)
# This mutator and initializator will take care of
# initializing valid individuals based on the allele set
# that we have defined before
genome.mutator.set(Mutators.G1DListMutatorAllele)
genome.initializator.set(Initializators.G1DListInitializatorAllele)
# Genetic Algorithm Instance
ga = GSimpleGA.GSimpleGA(genome)
ga.selector.set(Selectors.GRouletteWheel)
ga.setGenerations(800)
ga.stepCallback.set(evolve_callback)
# Do the evolution
ga.evolve()
# Best individual
print ga.bestIndividual()
def test_createAlleles_default(self):
_alleles = GAllele.GAlleles(allele_list=None)
self.assertTrue(hasattr(_alleles, 'allele_list'), True)
self.assertTrue(hasattr(_alleles, 'homogeneous'), True)
self.assertEqual(_alleles.allele_list, [])
_alleles = GAllele.GAlleles(allele_list=[1, 2, 3])
self.assertEqual(_alleles.allele_list, [1, 2, 3])
_alleles = GAllele.GAlleles(homogeneous=True)
self.assertEqual(_alleles.homogeneous, True)
def setUp(self):
self.alleles = GAllele.GAlleles()
self.alleles.add(MappedAlleleRange(10, 100))
self.alleles.add(MappedAlleleRange(0, 2, real=True))
self.alleles.add(MappedAlleleList([2, 4, 12]))
self.alleles.add(MappedAlleleList([-1]))
self.genome = G1DList.G1DList(len(self.alleles))
self.genome.setParams(allele=self.alleles, rangemin=-1, rangemax=1)
def test_AlleleRange_slicing(self):
_allelerange = GAllele.GAlleleRange(10, 20)
_allelerange.add(30, 40)
self.assertEqual(_allelerange[0], (10, 20))
_allelerange[1] = (50, 60)
self.assertEqual(_allelerange[1], (50, 60))
with self.assertRaises(ValueError):
_allelerange[1] = (60, 50)
def test_AlleleRange_add(self):
_allelerange = GAllele.GAlleleRange(10, 20)
_allelerange.add(30, 40)
self.assertEqual(_allelerange.beginEnd, [(10, 20), (30, 40)])
self.assertEqual(_allelerange.minimum, 10)
self.assertEqual(_allelerange.maximum, 40)
with self.assertRaises(ValueError):
_allelerange.add(40, 30)
def main_run(path):
global cm, coords
# write_random(filename, number of the cities, max width, max height)
#write_random("tsp_coords.txt", 50, 400, 400)
# load the tsp data file
filehandle = open("tsp_coords.txt", "r")
coords = path
cm = cartesian_matrix(coords)
# set the alleles to the cities numbers
setOfAlleles = GAllele.GAlleles(homogeneous=True)
lst = [ i for i in xrange(len(coords)) ]
a = GAllele.GAlleleList(lst)
setOfAlleles.add(a)
genome = G1DList.G1DList(len(coords))
genome.setParams(allele=setOfAlleles)
genome.evaluator.set(eval_func)
genome.mutator.set(Mutators.G1DListMutatorSwap)
genome.crossover.set(Crossovers.G1DListCrossoverOX)
genome.initializator.set(G1DListTSPInitializator)
ga = GSimpleGA.GSimpleGA(genome)
ga.setGenerations(300)
ga.setMinimax(Consts.minimaxType["minimize"])
ga.setCrossoverRate(1.0)
ga.setMutationRate(0.03)
ga.setPopulationSize(80)
#sqlite_adapter = DBAdapters.DBSQLite(identify="tsp", commit_freq=1000, frequency=500)
#ga.setDBAdapter(sqlite_adapter)
# This is to make a video
# ga.stepCallback.set(evolve_callback)
ga.evolve(freq_stats=100)
best = ga.bestIndividual()
best.fitness_list = fitness_list
if PIL_SUPPORT:
write_tour_to_img(coords,cm, best, "tsp_result_%d.png")
else:
print "No PIL detected, cannot plot the graph !"
return [best.getRawScore(), fitness_list]
def main_run(distancesFileName,
fragmentFileName,
crossover_rate=1.0,
mutation_rate=0.03,
population_size=80):
"""
@param distancesFileName String The file containing the pairwise distances of all
fragments
"""
global cm, coords, fragments
# Load the fragments
fragmentFile = open(fragmentFileName, "r")
fragments = readFragmentFile(fragmentFile)
# load the tsp data file
filehandle = open(distancesFileName, "r")
coords = read_coords(filehandle)
cm = coords
# set the alleles to the cities numbers
setOfAlleles = GAllele.GAlleles(homogeneous=True)
lst = [i for i in xrange(len(coords))]
a = GAllele.GAlleleList(lst)
setOfAlleles.add(a)
genome = G1DList.G1DList(len(coords))
genome.setParams(allele=setOfAlleles)
genome.evaluator.set(eval_func)
genome.mutator.set(Mutators.G1DListMutatorSwap)
genome.crossover.set(Crossovers.G1DListCrossoverOX)
genome.initializator.set(G1DListTSPInitializator)
ga = GSimpleGA.GSimpleGA(genome)
ga.setGenerations(5000) # 10000 is a good "real" value
ga.setMinimax(Consts.minimaxType["minimize"])
ga.setCrossoverRate(crossover_rate)
ga.setMutationRate(mutation_rate)
ga.setPopulationSize(population_size)
ga.evolve(freq_stats=100)
best = ga.bestIndividual()
return eval_func(best), best.getInternalList()
def run_ga(fast):
"""Initialize and run the genetic algorithms."""
alleles = GAllele.GAlleles() # create list of all alleles
alleles.add(GAllele.GAlleleRange(1, 1000)) # range for the number of days
alleles.add(GAllele.GAlleleRange(1, 20)) # range for hidden nodes
alleles.add(GAllele.GAlleleRange(1, 10)) # number of epochs
alleles.add(GAllele.GAlleleList([True, False])) # bias
alleles.add(
GAllele.GAlleleList([float(x * Decimal("0.1"))
for x in range(0, 10)])) # momentum
alleles.add(GAllele.GAlleleList([TanhLayer, SigmoidLayer])) # tanh/sigmoid
genome = G1DList.G1DList(len(alleles))
genome.setParams(allele=alleles, fast=fast)
genome.evaluator.set(eval_func)
genome.mutator.set(Mutators.G1DListMutatorAllele)
genome.initializator.set(Initializators.G1DListInitializatorAllele)
gen_alg = GSimpleGA.GSimpleGA(genome)
gen_alg.setMultiProcessing(not fast)
gen_alg.selector.set(Selectors.GRouletteWheel)
gen_alg.setGenerations(20)
gen_alg.evolve(freq_stats=0)
return gen_alg.bestIndividual()
def run_simulation(pset, max_generations, initial_list=None):
# Genome instance
setOfAlleles = GAllele.GAlleles()
setOfAlleles.add(GAllele.GAlleleRange(0, 51))
setOfAlleles.add(GAllele.GAlleleRange(0, 6))
setOfAlleles.add(GAllele.GAlleleRange(0, 51))
# Genome instance, 1D List of 50 elements
genome = G1DList.G1DList(3)
genome.setParams(allele=setOfAlleles)
#Fitness function
fe = FitnessEvaluator(pset.fitness_fnc, pset.fitness_temp)
Consts.CDefScaleLinearMultiplier = pset.fitness_temp
application = "OS" if "oxide" in pset.fitness_fnc.__name__ else "LS"
st = StatTrack(fe, pset.mutation_rate, pset.tournament_rate, include_ridiculous=pset.include_ridiculous, application=application)
genome.crossover.set(pset.crossover_fnc)
genome.mutator.set(pset.mutation_fnc)
genome.evaluator.set(fe.array_to_score)
genome.initializator.set(pset.initialization_fnc)
ga = GSimpleGA.GSimpleGA(genome)
ga.selector.set(pset.selection_fnc)
ga.terminationCriteria.set(AllFoundCriteria)
ga.stepCallback.set(st.evolve_callback)
ga.setPopulationSize(pset.popsize)
ga.setGenerations(max_generations)
if pset.elitism_num > 0:
ga.setElitism(True)
ga.setElitismReplacement((int)(math.ceil(pset.elitism_num * pset.popsize)))
else:
ga.setElitism(False)
ga.setMutationRate(pset.mutation_rate)
# TODO: figure out niching
stats_freq = 0
ga.evolve(freq_stats=stats_freq)
return st
def test_Alleles_slicing(self):
# includes slice operation, getitem and setitem
_alleles = GAllele.GAlleles(allele_list=[1, 2, 3])
self.assertEqual(_alleles[1], 2)
with self.assertRaises(Exception):
_ = _alleles[4]
_alleles[1] = 5
self.assertEqual(_alleles[1], 5)
self.assertEqual(_alleles[0:2], [1, 5])
def test_AlleleRange_getRandomAllele(self):
_allelerange = GAllele.GAlleleRange(10, 20)
random_allele = _allelerange.getRandomAllele()
self.assertTrue(
random_allele,
any([x[0] <= random_allele <= x[1]
for x in _allelerange.beginEnd]))
_allelerange.add(30, 40)
random_allele = _allelerange.getRandomAllele()
self.assertTrue(
random_allele,
any([x[0] <= random_allele <= x[1]
for x in _allelerange.beginEnd]))
_allelerange = GAllele.GAlleleRange(1.0, 2.0, real=True)
random_allele = _allelerange.getRandomAllele()
self.assertTrue(
random_allele,
any([x[0] <= random_allele <= x[1]
for x in _allelerange.beginEnd]))
def run_main():
# Genome instance
setOfAlleles = GAllele.GAlleles()
# From 0 to 10 we can have only some
# defined ranges of integers
for i in xrange(11):
a = GAllele.GAlleleRange(0, i)
setOfAlleles.add(a)
# From 11 to 19 we can have a set
# of elements
for i in xrange(11, 20):
# You can even add objects instead of strings or
# primitive values
a = GAllele.GAlleleList(['a', 'b', 'xxx', 666, 0])
setOfAlleles.add(a)
print setOfAlleles
genome = G1DList.G1DList(20)
genome.setParams(allele=setOfAlleles)
# The evaluator function (objective function)
genome.evaluator.set(eval_func)
# This mutator and initializator will take care of
# initializing valid individuals based on the allele set
# that we have defined before
genome.mutator.set(Mutators.G1DListMutatorAllele)
genome.initializator.set(Initializators.G1DListInitializatorAllele)
# Genetic Algorithm Instance
ga = GSimpleGA.GSimpleGA(genome)
ga.setGenerations(40)
# Do the evolution, with stats dump
# frequency of 10 generations
ga.evolve(freq_stats=5)
# Best individual
print ga.bestIndividual()
def __init__(self, generations, population, crossover, mutation, elite,
experiment, genCallBack, evalFunc, drivers):
self.experiment = experiment
self.population = population
self.crossoverProb = crossover
self.elite = elite
self.mutationProb = mutation
self.generations = generations
self.genCallBack = genCallBack
self.evalFunc = evalFunc
self.drivers = drivers
# sets pyevolve
# create alleles to make a GA model. each driver is represented as an allele
# each driver can take k different routes which is modelled by the different
# values each allele can take
driversAlleles = GAllele.GAlleles()
for dr in drivers:
lst = GAllele.GAlleleList(range(dr.od.numPaths))
driversAlleles.add(lst)
# define a genome with length = length of drivers
genome = G1DList.G1DList(len(drivers))
genome.setParams(allele=driversAlleles)
genome.evaluator.set(self.evalFuncCallback)
genome.initializator.set(self.initGenome)
# Genetic Algorithm Instance
self.ga = GSimpleGA.GSimpleGA(genome)
self.ga.setMinimax(Consts.minimaxType["minimize"])
self.ga.selector.set(Selectors.GRankSelector)
self.ga.setGenerations(self.generations)
self.ga.stepCallback.set(self.genCallBack)
self.ga.setMutationRate(self.mutationProb)
self.ga.setCrossoverRate(self.crossoverProb)
self.ga.setPopulationSize(self.population)
self.ga.terminationCriteria.set(GSimpleGA.RawStatsCriteria)
self.ga.setElitism(True)
self.ga.setElitismReplacement(self.elite)
self.ga.setSortType(Consts.sortType["raw"])
def run(self):
# Allele define valid chromosome value
alleles = GAllele.GAlleles()
# Define gene with 2 chromosomes
# MA type
alleles.add(GAllele.GAlleleList([0, 1, 2, 3, 4]))
# MA range
alleles.add(GAllele.GAlleleRange(1, 99))
# Genome instance, 1D List
genome = G1DList.G1DList(len(alleles))
# Sets the range max and min of the 1D List
genome.setParams(allele=alleles)
# The evaluator function (evaluation function)
genome.evaluator.set(self.fitness)
# This mutator and initializator will take care of
# initializing valid individuals based on the allele set
# that we have defined before
genome.mutator.set(Mutators.G1DListMutatorAllele)
genome.initializator.set(Initializators.G1DListInitializatorAllele)
# Genetic Algorithm Instance
ga = GSimpleGA.GSimpleGA(genome)
# Set the Roulette Wheel selector method, the number of generations and
# the termination criteria
ga.selector.set(Selectors.GRouletteWheel)
ga.setGenerations(self.gen)
ga.setPopulationSize(self.pop)
ga.terminationCriteria.set(GSimpleGA.ConvergenceCriteria)
pop = ga.getPopulation()
pop.scaleMethod.set(Scaling.SigmaTruncScaling)
ga.evolve(freq_stats=10)
# Best individual
self.best = ga.bestIndividual()
def run_main(): print "Starting main()!" #sys.stdout.flush() global NUM_GENS, NUM_POP, NUM_ELITE, GLOB_COUNT, NO_OF_PARAMS setOfAlleles = GAllele.GAlleles() sf = GAllele.GAlleleList(strp_fac) su = GAllele.GAlleleList(strp_unt) cn = GAllele.GAlleleList(cb_nds) # cs = GAllele.GAlleleList(cb_buf_size) al = GAllele.GAlleleList(alignment) sb = GAllele.GAlleleList(siv_buf_size) setOfAlleles.add(sf) setOfAlleles.add(su) setOfAlleles.add(cn) # setOfAlleles.add(cs) setOfAlleles.add(al) setOfAlleles.add(sb) genome = G1DList.G1DList(NO_OF_PARAMS) genome.setParams(allele=setOfAlleles) genome.evaluator.set(eval_func) genome.mutator.set(Mutators.G1DListMutatorAllele) genome.initializator.set(Initializators.G1DListInitializatorAllele) ga = GSimpleGA.GSimpleGA(genome) ga.selector.set(Selectors.GRouletteWheel) ga.setMutationRate(0.15); ga.setGenerations(NUM_GENS) #ga.terminationCriteria.set(ConvergenceCriteria) ga.setPopulationSize(NUM_POP) ga.setMinimax(Consts.minimaxType["minimize"]) ga.setElitism(True) ga.setElitismReplacement(NUM_ELITE) print 'ga.evolve' ga.evolve(freq_stats=1) print 'closing result' result_output.close() print 'Best solution' print ga.bestIndividual()
list_ += value
return list_
def convert_to_dict(chromosome):
dict_ = {}
keys = init_cutpoints.keys()
i = 0
for key in keys:
dict_[key] = chromosome[i:i+3]
i += 3
return dict_
chromosome_ = convert_to_list(init_cutpoints)
setOfAlleles = GAllele.GAlleles()
keys = init_cutpoints.keys()
for key in keys:
for _ in range(0, 3):
if key != 'cip':
a = GAllele.GAlleleRange(int(df[key].min()), int(df[key].max()))
setOfAlleles.add(a)
else:
a = GAllele.GAlleleRange(0, 10)
setOfAlleles.add(a)
df_actual = pd.read_csv("data/Cust_Actual.csv", index_col=['index'])
genome = G1DList.G1DList(len(chromosome_))
genome.setParams(allele=setOfAlleles)
def test_AlleleList_getRandomAllele(self):
_allelelist = GAllele.GAlleleList(options=[1, 2, 3])
random_allele = _allelelist.getRandomAllele()
self.assertIn(random_allele, _allelelist.options)
def run_ga():
global ga
#___________________Genome instance
#
setOfAlleles = GAllele.GAlleleList()
pars_min = velocity_min + depth_min + vpvs_min
pars_max = velocity_max + depth_max + vpvs_max
num_pars = len(pars_min)
for (vmin, vmax) in zip(pars_min, pars_max):
tmp = GAllele.GAlleleRange(vmin, vmax, real=True)
setOfAlleles.add(tmp)
genome = G1DList.G1DList(num_pars)
genome.setParams(allele=setOfAlleles)
genome.initializator.set(Initializators.G1DListInitializatorAllele)
genome.mutator.set(Mutators.G1DListMutatorAllele)
genome.crossover.set(Crossovers.G1DListCrossoverUniform)
#___________________The evaluator function (objective function)
#
genome.evaluator.set(eval_func)
#___________________Genetic Algorithm Instance
#
ga = GSimpleGA.GSimpleGA(genome, seed=int(seed))
if num_cpu: ga.setMultiProcessing(True, True, int(num_cpu))
if ga_selector == 'T': ga.selector.set(Selectors.GTournamentSelector)
if ga_selector == 'R': ga.selector.set(Selectors.GRouletteWheel)
if ga_selector == 'N': ga.selector.set(Selectors.GRankSelector)
if ga_selector == 'M':
ga.selector.setRandomApply(True)
ga.selector.set(Selectors.GTournamentSelector, 0.75)
ga.selector.add(Selectors.GRouletteWheel, 0.20)
ga.selector.add(Selectors.GRankSelector)
ga.setMinimax(Consts.minimaxType["minimize"])
ga.setGenerations(int(generationSize))
ga.setPopulationSize(int(populationSize))
ga.setCrossoverRate(pCrossover)
ga.setMutationRate(pMutation)
ga.setElitism(True)
ga.setElitismReplacement(int(num_etlsm))
#___________________Sets the DB Adapter
#
sqlite_adapter = DBAdapters.DBSQLite(identify=dbase_name,
resetDB=eval(resetDB))
ga.setDBAdapter(sqlite_adapter)
#___________________Do the evolution
#
ga.evolve(freq_stats=5)
#___________________Print Best individual
#
best = ga.bestIndividual()
best_rs = best.getRawScore()
best_v = best.genomeList[:len(velocity_min)]
best_d = best.genomeList[len(velocity_min):2 * len(velocity_min)]
best_r = best.genomeList[2 * len(velocity_min):]
print ''
print '+++ Best Raw Score =', best_rs
print '+++ FinalModel :'
print ' +++ Velocity :', rnd(best_v, 2)
print ' +++ Depth :', rnd(best_d, 2)
print ' +++ VpVs :', rnd(best_r, 2)
return best, best_rs, best_v, best_d, best_r
#corra este archivo p4.py cada vez que cambie Pmut, Pcros o Npop, #el programa guardara los datos de fitness maximoen cada generacion de dicha evolucion en el archivo 'fitp4.txt' #El archivo 'fitp4.txt' se reescribe cada vez que corre la evolucion, asi que guardelo para cada valor que cambie #Puede hacer los graficos en Matlab #use 3 valores distintos para la probabilidad de mutacion, 3 distintos para la probabilidad de cruzamiento, #y 3 distintos numero de individuos por poblacion Pmut=0.03 #probabilidad de mutacion Pcros=0.9 #probabilidad de cruzamiento Npop=80 #Numero de individuos por poblacion #Elija el numero de generaciones que quiera Numero_generaciones= setOfAlleles = GAllele.GAlleles() genome = G1DList.G1DList() genome.evaluator.set() genome.mutator.set() genome.crossover.set() genome.initializator.set(G1DListTSPInitializator) ga = GSimpleGA.GSimpleGA(genome) ga.selector.set() ga.setGenerations(Numero_generaciones) ga.setCrossoverRate(Pcros) ga.setMutationRate(Pmut) ga.setPopulationSize(Npop)
def test_AlleleList_add(self):
_allelelist = GAllele.GAlleleList(options=[1, 2, 3])
_allelelist.add(4)
self.assertEqual(_allelelist.options, [1, 2, 3, 4])
def test_AlleleList_slicing(self):
_allelelist = GAllele.GAlleleList(options=[1, 2, 3])
self.assertEqual(_allelelist[0:2], [1, 2])
self.assertEqual(_allelelist[1], 2)
_allelelist[1] = 4
self.assertEqual(_allelelist[1], 4)
def Grid_Constructor(numline, numParm, data):
alleles = GAllele.GAlleles()
for i in range(0, numline * numParm):
alleles.add(GAllele.GAlleleRange(data[i][0], data[i][1]))
return alleles
import time
time.ctime()
import re
time_label = re.sub(r'[^0-9]', "_", str(dt.datetime.now())) + "_Case2"
time_label
# In[ ]:
Start = dt.datetime.now()
S = dt.datetime.now()
genome = G1DList.G1DList(len(gd_req))
setOfAlleles = GAllele.GAlleles()
for i in np.array(gd_req["skill_requirement"]):
# You can even add an object to the list
try:
b = skill_matrix.loc[i].unique()
except:
b = [0]
a = GAllele.GAlleleList(b)
setOfAlleles.add(a)
genome.setParams(allele=setOfAlleles)
genome.mutator.set(mutator_allel_final)
#genome.initializator.set(Initializators.G1DListInitializatorAllele)
genome.initializator.set(initializator_final_temp)
genome.crossover.set(crossover_uniform_final_all)
lambd = chromo[2]
eta = chromo[3]
rho = chromo[4]
diffs = np.zeros(len(strikes) * len(terms))
n = 0
for i, k in enumerate(strikes):
for j, t in enumerate(terms):
diffs[n] = vanilla[i, j] - _heston.ucall(futs[j], k, t, v, vbar,
lambd, eta, rho)
n += 1
score = norm(diffs)
return score
# Genome instance
setOfAlleles = GAllele.GAlleles()
setOfAlleles.add(GAllele.GAlleleRange(0.0, 5.0, True))
setOfAlleles.add(GAllele.GAlleleRange(0.0, 5.0, True))
setOfAlleles.add(GAllele.GAlleleRange(5.0, 15.0, True))
setOfAlleles.add(GAllele.GAlleleRange(0.0, 5.0, True))
setOfAlleles.add(GAllele.GAlleleRange(-1, 1, True))
genome = G1DList.G1DList(5)
genome.setParams(allele=setOfAlleles)
genome.evaluator.set(heston_evaluate)
genome.mutator.set(Mutators.G1DListMutatorAllele)
genome.initializator.set(Initializators.G1DListInitializatorAllele)
# Genetic Algorithm Instance
ga = GSimpleGA.GSimpleGA(genome)
# ga.setGenerations(100)
def generateDecision(self):
setOfAlleles = GAllele.GAlleles(homogeneous=True)
#raw_input("Press ENTER to exit")
# prepare indices
#print "Preparing indices..."
indicesContainer = IndicesContainer(self.training_low_prices,
self.training_high_prices,
self.training_close_prices,
self.training_volumes)
indicesContainer.computeIndices()
indicesList = indicesContainer.getIndicesList()
# prepare sell trend and buy trend vectors
#print "Preparing sell trend and buy trend vectors..."
#normalizedIndicesList = indices.indicesNormalizer().normalize(indicesContainer.getIndicesList())
tradingGA.sellTrendBeginning, tradingGA.buyTrendBeginning = simpleTrendBeginningsFinder(
self.training_close_prices, indicesList).findTrendBeginnings()
# prepare individual
#print "Preparing first individual..."
a = GAllele.GAlleleList(indicesList)
setOfAlleles.add(a)
genome = G1DList.G1DList()
genome.setParams(allele=setOfAlleles)
genome.evaluator.set(eval_func)
genome.mutator.set(G1DListStockMutator)
genome.crossover.set(G1DListStockCrossoverOperator)
genome.initializator.set(G1DListStockInitializator)
# prepare engine
#print "Preparing Genetic Algorithm engine..."
ga = GSimpleGA.GSimpleGA(genome)
ga.setGenerations(self.generations)
ga.setMinimax(Consts.minimaxType["maximize"])
ga.setCrossoverRate(1.0)
ga.setMutationRate(0.1)
ga.setElitismReplacement(3)
ga.selector.set(Selectors.GUniformSelector)
ga.setPopulationSize(self.population_size)
#print "Executing indices subset search..."
ga.evolve(freq_stats=0)
best = ga.bestIndividual()
#print "Generating trading signal..."
#print "Preparing prediction set indices..."
predictionIndicesContainer = IndicesContainer(
self.prediction_low_prices, self.prediction_high_prices,
self.prediction_close_prices, self.prediction_volumes)
predictionIndicesContainer.computeIndices()
predictionIndicesList = predictionIndicesContainer.getIndicesList()
# prepare sell trend and buy trend vectors
#print "Preparing prediction set sell trend and buy trend vectors..."
sell = computeClusterCentre(best.getInternalList(),
tradingGA.sellTrendBeginning)
buy = computeClusterCentre(best.getInternalList(),
tradingGA.buyTrendBeginning)
predictionIndicesList = pickIndices(predictionIndicesList,
best.getInternalList())
prediction = computeClusterCentre(
predictionIndicesList, len(predictionIndicesList[0].getResult(
))) #trend beginning to ostatni element listy z danymi
d1 = numpy.linalg.norm(numpy.asarray(buy) - numpy.asarray(prediction))
d2 = numpy.linalg.norm(numpy.asarray(sell) - numpy.asarray(prediction))
if d1 < d2:
#print "Time for Buyin'"
return 1
elif d1 > d2:
#print "Time for Sellin'"
return -1
return 0