def __init__(this, gene=None):
super(onemax_ptype, this).__init__(gene)
if this.gene == None:
this.generate_gene(problem_size)
problem = raw_input("Input problem to solve (matching/onemax): ")
if problem == "matching":
target = [int(c) for c in raw_input("Input a bit string: ")]
problem_size = len(target)
else:
problem_size = int(input("Input problem size: "))
target = [1 for i in xrange(problem_size)]
population_list = evoalg.main(onemax_ptype)
best = [pop.max_fitness() for pop in population_list]
average = [pop.average_fitness() for pop in population_list]
average_plus_stddev = [
pop.average_fitness() + pop.fitness_standard_deviation()
for pop in population_list
]
average_minus_stddev = [
pop.average_fitness() - pop.fitness_standard_deviation()
for pop in population_list
]
std_dev = [pop.fitness_standard_deviation() for pop in population_list]
print "Plotting..."
fill_between(range(len(best)),
def __init__(this, gene=None):
super(onemax_ptype, this).__init__(gene)
if this.gene == None:
this.generate_gene(problem_size)
problem = raw_input("Input problem to solve (matching/onemax): ")
if problem == "matching":
target = [int(c) for c in raw_input("Input a bit string: ")]
problem_size = len(target)
else:
problem_size = int(input("Input problem size: "))
target = [1 for i in xrange(problem_size)]
population_list = evoalg.main(onemax_ptype)
best = [pop.max_fitness() for pop in population_list]
average = [pop.average_fitness() for pop in population_list]
average_plus_stddev = [pop.average_fitness() + pop.fitness_standard_deviation() for pop in population_list]
average_minus_stddev = [pop.average_fitness() - pop.fitness_standard_deviation() for pop in population_list]
std_dev = [pop.fitness_standard_deviation() for pop in population_list]
print "Plotting..."
fill_between(
range(len(best)), average_plus_stddev, average_minus_stddev, alpha=0.2, color="b", label="Standard deviation"
)
plot(range(len(best)), best, color="r", label="Best")
plot(range(len(best)), average, color="b", label="Average with std. dev.")
title("Fitness plot - Onemax")
xlabel("Generation")
def __init__(this, gene=None):
super(blotto_ptype, this).__init__(gene)
if this.gene == None:
this.generate_gene(problem_size*4)
this.generate_phenotype()
def calc_entropy(this):
return -sum((battle*log(battle, 2) if battle != 0 else 0) for battle in this.phenotype)
problem_size = int(input("Input number of battles: "))
rf = float(input("Input reployment fraction: "))
lf = float(input("Input \"morale reduction from loss\"-fraction: "))
population_list = evoalg.main(blotto_ptype)
for pop in population_list:
pop.sort()
average_entropy = [sum(strategy.calc_entropy() for strategy in pop.get_individuals())/problem_size for pop in population_list]
best_entropy = [pop.get_individuals()[0].calc_entropy() for pop in population_list]
best = [pop.max_fitness() for pop in population_list]
average = [pop.average_fitness() for pop in population_list]
std_dev = [pop.fitness_standard_deviation() for pop in population_list]
worst = [pop.min_fitness() for pop in population_list]
topline = [average[i] + std_dev[i] for i in xrange(len(population_list))]
bottomline = [average[i] - std_dev[i] for i in xrange(len(population_list))]
figure(1)
def __init__(this, gene=None):
super(blotto_ptype, this).__init__(gene)
if this.gene == None:
this.generate_gene(problem_size * 4)
this.generate_phenotype()
def calc_entropy(this):
return -sum((battle * log(battle, 2) if battle != 0 else 0)
for battle in this.phenotype)
problem_size = int(input("Input number of battles: "))
rf = float(input("Input reployment fraction: "))
lf = float(input("Input \"morale reduction from loss\"-fraction: "))
population_list = evoalg.main(blotto_ptype)
for pop in population_list:
pop.sort()
average_entropy = [
sum(strategy.calc_entropy()
for strategy in pop.get_individuals()) / problem_size
for pop in population_list
]
best_entropy = [
pop.get_individuals()[0].calc_entropy() for pop in population_list
]
best = [pop.max_fitness() for pop in population_list]
average = [pop.average_fitness() for pop in population_list]
std_dev = [pop.fitness_standard_deviation() for pop in population_list]
