def firstTrial():
# create LocatedModels and add them into populationArray[]
for i in range(sizeOfPopulation):
locatedModel = MODEL.LocatedModel(
numberOfNodes) # create default locatedModel
# define 1 chromosomes in DNA of current locatedModel randomly untill ones is equal to the pmedian
counter = 0
while counter < pmedian:
randomIndex = randint(0, len(locatedModel.DNA) - 1)
if (locatedModel.DNA[randomIndex] is not 1):
locatedModel.DNA[randomIndex] = 1
counter += 1
FitnessLevelFinder(
locatedModel
) # calculate fitness value for the current locatedModel
locatedModelList.append(
locatedModel
) # after all of properties of the current locatedModel have been defined add into locatedModelList[]
def newLocatedModel(dad, mom):
baby = MODEL.LocatedModel(numberOfNodes) # create a default baby
# UNIFORM CROSSOVER chromosomes are chosen radomly form dad and mom
for i in range(len(baby.DNA)):
randomChromosome = randint(0, 1)
if randomChromosome is 0:
baby.DNA[i] = dad.DNA[i]
else:
baby.DNA[i] = mom.DNA[i]
# In this part fix amount of ones in DNA if there are more or less than pmedian
onesIndexes = baby.getOnes()
difference = pmedian - len(
onesIndexes
) # difference<0 means number of ones more than pmedian so we need to remove some ones randomly
if difference < 0:
for i in range(abs(difference)):
r = randint(0, len(onesIndexes) - 1)
randomDnaIndex = onesIndexes[r]
baby.DNA[randomDnaIndex] = 0
onesIndexes.remove(onesIndexes[r])
elif difference > 0:
dnaLength = len(baby.DNA)
i = 0
while i is not difference:
randomDnaIndex = randint(0, dnaLength - 1)
if (baby.DNA[randomDnaIndex] is 0):
baby.DNA[randomDnaIndex] = 1
i += 1
else:
pass
mutation(baby) # check mutation factor for baby
FitnessLevelFinder(baby) # get fitness value of baby
return baby
