def TTCvsMMMD(gridPoint):
if not (gridPoint['compositeClass0']
== 'integerVectorSymmetricRangeMutations'
and gridPoint['probabilityOfMutatingClass0'] == 0.2):
return
# determining the time of convergence
bilogFileName = 'bestIndividual{}.log'.format(gridPoint['randomSeed'])
bilog = np.loadtxt(bilogFileName)
toc = None # time of convergence
# TOME OF CONVERGENCE
for i in range(bilog.shape[0]):
if -1. * bilog[i, 1] <= fitnessThreshold:
toc = i
break
if toc is None:
with open('../results/nonconverged runs', 'a') as ncrf:
ncrf.write(str(gridPoint['randomSeed']) + '\n')
return
# making a scatter of points of TTC vs MMMD
with open('../results/ttcvsmmmd', 'a') as tmfile:
for i in range(evsDefaults['genStopAfter'] + 1):
tmfile.write('{} {}\n'.format(
toc - i, gctools.minParetoFrontHammingDistanceToMMM(i)))
def generateMinMMMDistTimeSeries(gridPoint):
minMMMDistTS = [
gctools.minParetoFrontHammingDistanceToMMM(gen)
for gen in range(1, evsDefaults['genStopAfter'] + 1)
]
filename = minMMMDistFileName(gridPoint)
with open(filename, 'a') as file:
file.write(' '.join(map(str, minMMMDistTS)) + '\n')
def generateMinMMMDistTimeSlices(gridPoint):
return [
gctools.minParetoFrontHammingDistanceToMMM(gen) for gen in stages
]